Pus in the pleura, the pericardium, and the peritoneum
From Primary Surgery
Pus in the pleural cavities[md]empyema
Pus usually reaches a patient's pleural cavity from infection of the lung under it. This can be pneumonia, a lung abscess, or the pneumonitis that may follow an inhaled foreign body (usually in a child), or carcinoma of the bronchus (usually in a cigarette smoker). Occasionally, an empyema is tuberculous; rarely it may follow rupture of a liver or subphrenic abscess through the diaphragm.
A common history is that a week or more ago, as the patient was beginning to recover from a chest infection, improvement stopped. He now remains ill, anorexic and febrile, and is starting to lose weight[md]despite antibiotics.
Many kinds of bacteria can be responsible, but pneumococci are perhaps the most common. Antibiotics are only effective in the earliest stages, and may mask the symptoms of an empyema later. The result is that empyemas can remain undetected for years and are often missed in a busy outpatient department. This is sad because you can treat them, so watch for them, and ask your staff to do so too.
Pus in the pleural cavity, like pus anywhere else, must be removed. To begin with it is thin, like serum; later it thickens and looks like scrambled egg. So adapt your method of removing it to its thickness. While it is still thin, aspirate it with a syringe and needle. When it is too thick for this, but is still fluid enough to flow down a tube into a bottle, use closed drainage, as if you were draining blood from an injured chest (65.2). The surfaces of the patient's pleura will not have stuck together at this stage, so you will have to use an underwater seal to prevent air getting into his pleural cavity and letting his lung collapse.
If the pus in his pleural cavity is left undrained, it will soon become too thick to flow down a long thin tube into a bottle. Once his empyema has reached this stage, only thoracotomy and decortication will properly expand his lung, but this is a dangerous operation, even in expert hands. If referral is difficult, you can improve him greatly by draining pus through an open drain. To do this you will have to remove a piece of a rib and push a short wide tube through its bed. The surfaces of his pleura will now have stuck so firmly that air cannot enter his pleural cavity to collapse his lung. When you do this be sure to: (1) Remove the piece of rib from inside its periosteum, so as not to injure the vessels and nerve which run just below it. (2) Place the inner end of the tube where the bottom of the abscess cavity will be while he sits in his usual position in bed.
If the pus in his pleural cavity remains even longer it will be replaced by fibrous tissue which will be very difficult to remove.
Children have special problems. In a child an empyema may follow a post-measles pneumonia, or the rupture of a staphylococcal lung abscess into a pleural cavity. He is likely to be between 1 and 3 years, malnourished, anaemic, and anorexic, with a persistent cough, fever, dyspnoea, diarrhoea, and perhaps vomiting. He may be very sick indeed with a pyopneumothorax under tension. Treat him as you would an adult.
ASPIRATOR, Martin's, with 3-way tap and needles, one only. This is much the best instrument for draining large quantities of fluid from the chest. It has a tap, so that you can aspirate pus and discharge it through a tube into a receiver, without letting air enter the chest.
Fig. 6-1 THE ANATOMY OF THE PLEURAE. A, the relation of a patient's pleurae and lungs to his chest wall. B, a coronal section of his thorax (semischematic). C, the ventral aspect of his thorax showing the surface projections of his heart and pleurae. D, the subdivisions of his mediastinum.
1, the horizontal fissure. 2, the oblique fissure. 3, the inferior border of the right lung. 4, the costodiaphragmatic reflexion of the right pleura. 5, the costodiaphragmatic reflexion of the left pleura. 6, the cardiac notch of the left lung. A, Fig. 8.23B and C, Fig. 8.23A from Gray's Anatomy' (Churchill Livingstone). B, Fig. 6.6 and D, Fig. 6.8 from Grant's Method of Anatomy' (9th edition 1975 edited by JV Basmajian). With kind permission.
EMPYEMA CLINICAL FEATURES. If an empyema involves the whole of a patient's pleural cavity and contains a litre or more of pus, you should be able to diagnose it clinically. Look for limited movement of his chest on the affected side, shifting of his trachea and apex beat, dullness to percussion, reduced breath sounds and reduced vocal fremitus. Vocal resonance (the sound 99) may be high-pitched at the top of the empyema and absent over its lower part.
X-RAYS. usually show a dense area at one lung base. Take a PA and a lateral to show the site and extent of the empyema.
ANTIBIOTICS. When an empyema is established, antibiotics are ineffective. Pus must be drained. If he has fever or malaise give him chloramphenicol (2.7) until sensitivity tests show the need for change.
ASPIRATING [s7]A PLEURAL EFFUSION INDICATIONS. (1) To confirm the diagnosis. (2) To remove the bulk of the fluid in the early stages while it is still thin.
EQUIPMENT. A Martin's aspirator, with its needles and 3- way tap, a 20 ml syringe, local anaesthetic solution and a receiver. Or, improvise the equipment in Fig. 65-8.
METHOD. Premedicate the patient thoroughly an hour before. If he is not very ill, sit him astride a chair leaning over a pillow on the backrest. If he is very ill, sit him in bed with his arms folded, leaning over a bed table or a pile of pillows.
Aspirate near the lowest point of the empyema, as defined on the PA and lateral X-rays. To establish this, aspirate several sites if necessary, so as to find the lowest site that yields pus, but remember the surface markings of the pleura (9-4a). Commonly, the posterior axillary line is the right vertical line in which to aspirate.
Infiltrate anaesthetic solution into his skin and subcutaneous tissues over the chosen space, and also a space above and below[md]you may have chosen the wrong one.
Insert the needle, pierce his pleura and aspirate gently; turn the tap and discharge the fluid into a receiver. If you don't have a 3-way tap, and have not improvised the equipment in Fig. 65-8, you can (less desirably) put your finger over the hub of the needle, as you disconnect the syringe to discharge it.
Repeat the aspiration 2 or 3 times a week until pus stops forming, or it becomes too thick to aspirate.
CLOSED DRAINAGE [s7]FOR A PLEURAL EFFUSION Many empyemas do not resolve on aspiration alone. If pus thickens, so that aspiration is even a little difficult, closed drainage is necessary. Insert an underwater seal drain, as for a haemothorax (65.2). Leave it for a least two weeks until firm adhesions have formed between the surfaces of his pleura, which will prevent his lung collapsing when you take the tube out. The instillation of 5 to 10 g of lipiodol before repeat X-rays is a useful way of defining the lowest point of the empyema.
If he improves, and X-rays show disappearance of the empyema and re-expansion of his lung, cut the stitch securing the tube, pull it out and quickly press an airtight dressing over the hole.
If he does not improve, he needs open drainage or referral.
Fig 6-1a STAGES IN THE DRAINAGE OF PUS IN THE PLEURA. A, a very recent pleural effusion can be drained with a syringe and needle. B, if pus becomes too thick for this, you will have to use a rubber tube and an underwater seal drain in a bottle (closed drainage). C, if pus becomes even thicker, resect the patient's rib, and insert a short wide tube (open drainage). Shorten this tube as his empyema drains, and make sure it is in the bottom of the cavity. D, if you fail to drain an empyema, you may have to refer the patient for decortication.
OPEN DRAINAGE [s7]FOR AN EMPYEMA, RESECTING A RIB INDICATIONS. Draining an empyema when closed drainage has failed. The patient's lung must have stuck to his ribs. (1) The traditional pre-antibiotic test was to put some of the fluid in a test tube; if the sediment was approximately half the volume of the fluid, it was safe to insert an open drain. Antibiotics make this test less useful, because the fluid is more likely to remain thin. (2) Slowly withdraw the tube of the underwater seal drain from the water. If the column of water does not run up towards the pleura, but stays in the tube, his pleura has stuck to his ribs, so that an an underwater seal is unnecessary and open drainage can start.
CONTRAINDICATIONS. A tuberculous empyema. See below under empyema necessitans.
X-RAYS. Examine PA and lateral views with the greatest care to see which rib to resect. If you cannot easily see the lowest point of an empyema, inject 10 ml of oily contrast medium before you take the films.
ANAESTHESIA. Take him to the theatre. Premedicate him well (A 5.2). Use a combination of local infiltration (A 5.4) and intercostal blocks (A 6.7). Block his intercostal nerves at the site of your chosen incision, and also one rib above and one below it as far back as possible.
METHOD. Drain his empyema from its lowest point, in the position he would be in while he sits in bed. Since he lies more supine than prone, choose the lowest point of the empyema posteriorly. Often, his 9th rib in the paravertebral line is the best, but it may be below this.
CAUTION ! Don't make the opening too low, because his diaphragm will rise as the pus drains and block the opening. It should always be at least one space abve his diaphragm.
Sit him on a stool leaning forwards against the operating table. Before incising, confirm by aspiration through more than one intercostal space, that you have chosen the correct rib to remove. Make an 8 cm vertical incision, extending above and below the selected rib, so that you can more easily resect the rib on either side if necessary.
Cut down to the rib, and incise the periosteum along its centre. Use a curved Faraboef rougine to strip the periosteum with its attached intercostal muscles from the outer surface of his rib. Clean its upper and lower borders. Then use Doyen's raspatory (or Faraboef's rougine) to remove the periosteum from its inner surface. Strip its upper and lower borders as in Fig 6-2.
CAUTION ! (1) The intercostal blocks should have anaesthetized his parietal pleura adequately; if they have not, repeat the intercostal blocks and wait. If you fail to anaesthetize him adequately, extreme pain may cause vasovagal shock. (2) His intercostal vessels can bleed severely if you fail to identify them, so be sure to avoid them by keeping inside the periosteum.
Excise a 5[nd]10 cm length of rib with an osteotome, rib shears, or a large pair of bone cutters. Make an incision in the bed of this rib through into his pleural cavity. Open it with a haemostat, explore it it with your finger, and remove what semisolid pus you can with sponge holders. He will probably start coughing.
CAUTION ! (1) If when you explore the cavity with your finger, you find that you have not removed the rib at the bottom of the cavity, remove the rib below. If you don't do this his empyema will not resolve completely. (2) Send pus for smear and culture, it may be tuberculous; tuberculous pus looks different, is more watery, contains particles and should not be drained anyway (see above and below).
Fix a wide tube in the empyema cavity, leaving about 2 cm above the skin surface. Fix it with a suture, a safety pin and adhesive strapping; apply a large gauze and cotton wool dressing.
POSTOPERATIVELY, encourage him to do vigorous breathing exercises. Monitor the size of the cavity by introducing contrast medium and taking X-rays. Alternatively, measure how much sterile saline you can run into it.
When drainage stops or becomes less than 5 ml/day, remove the tube. The residual sinus will heal, provided that there is no bronchopleural fistula. This can take 2 or 3 months.
Fig. 6-2 RESECTING A RIB. A, the patient's empyema covered with a thick layer of fibrous tissue. B, a common site for draining an empyema[md]his 9th rib in his paravertebral line. Vary this as the occasion demands. C, his skin incised, showing the incision over the periosteum. D, reflecting the periosteum with Faraboef's rougine. E, reflecting the periosteum off the inner surface of his rib. F, completing the task with Doyen's raspatory. G, resecting the rib. H, preparing to incise the periosteum in the bed of the rib. I, sucking out the pus. J, and K, putting in a finger to break down the loculi. L, a drainage tube in place.
CHILDREN [s7]WITH EMPYEMAS You cannot drain a small child's pleural cavity adequately by inserting an intercostal drain between two ribs, because the drain will be nipped by his ribs or obstructed by pus. So remove a centimetre or two of rib, using ketamine, to make a hole which is big enough for a tube. Adequate drainage will eventually cure him if: (1) his lung is not immobilized with thick fibrin, (2) he has no bronchopleural fistula, and (3) his empyema is localized.
Start drainage with an underwater seal drainage bottle. This will limit his activity, and may cause the drain to be pulled out; but his lung will expand. If necessary, drain the cavity with another high pleural drain as for a pneumothorax, see Fig. 65-2.
When you are confident that his lung has stuck to his ribs (see above), cut the tube short, fit it with a pin and butterfly strapping, put a colostomy bag over it to collect the pus and allow him up. Increased activity is the best physiotherapy. If he does not settle in 3 or 4 weeks, refer him.
DIFFICULTIES [s7]WITH AN EMPYEMA If his EMPYEMA PRESENTS ON HIS CHEST WALL (EMPYEMA NECESSITANS, unusual), it is almost sure to be tuberculous, perhaps a complication of AIDS. The signs suggestive of a tuberculous empyema are: (1) Swelling of the chest wall, starting first with swelling of its intercostal spaces. (2) Sinuses. (3) X-ray signs of pulmonary tuberculosis. (4) Typically, fluid on aspiration which is not pus, but is thin and watery with small particles of necrotic tissue. You may or may not find AAFB in the smear. Don't drain it or it will become secondarily infected. Give him chemotherapy for tuberculosis. If his mediastinum is shifted (unusual), aspirate some fluid.
If AIR COMES OUT WITH THE PUS, he has a BRONCHOPLEURAL FISTULA which is unlikely to close spontaneously. When his condition has improved, refer him to an expert thoracic surgeon, who will find the task of closing the fistula difficult.
If a patient with tuberculosis has an AIR-FLUID LEVEL IN A PLEURAL CAVITY, he has a TUBERCULOUS BRONCHOPLEURAL FISTULA. Give him chemotherapy (29.1). Using careful aseptic precautions, insert an underwater seal drain (65.2) and leave it in until his lung has expanded.
If his INTERCOSTAL VESSELS BLEED, encircle them with a needle and thread. Avoid tying the nerve because this is painful. If you have difficulty, transfix them with a ligature, so that they are compressed against the stump of the rib which remains.
If his EMPYEMA FAILS TO HEAL: (1) You may have put the drainage tube too high or too far forward. (2) You may have removed it too early. (3) You may have put it in too late. (4) There may be a foreign body, such as a piece of drainage tube, in his chest. (5) He may have developed a fistula between his bronchi and his pleura. (6) He may have tuberculosis, carcinoma, actinomycosis, or a ruptured amoebic liver abscess. Further dependent drainage is all that he probably needs for (1), (2) or (3). Instil 5 to 10 ml of contrast medium, repeat the X-ray, and if necessary resect another rib. If this fails, refer him.
Pus in the pericardium
Fluid sometimes accumulates in a patient's pericardium. If there is only a little, you can leave it there (if indeed you are aware of it at all), unless you need it for diagnosis. But if there is much, it embarrasses the action of his heart (cardiac tamponade) and may kill him, so you may have to remove it urgently! The fluid can be blood after a cardiac injury (65.9) or an effusion from many causes, either infected or sterile. Elsewhere in the body, you drain pus to treat an infection. In his pericardium, you are mainly draining it to overcome its mechanical effects.
He is unlikely to present with symptoms that immediately suggest a pericardial effusion. He is more likely to be admitted with a variety of medical diseases and be observed to have some of the following signs: (1) Grossly distended neck veins, (2) pulsus paradoxus (a reduction of arterial pressure of [mt] 10 mm and of pulse pressure on inspiration), (3) pulsus alternans (QRS complexes of alternately varying voltage), and (4) a large cardiac shadow. He is obviously ill', and may be febrile. He has signs of a low cardiac output with a poor peripheral circulation; he has a small pulse volume, tachycardia, a low normal or subnormal blood pressure, and soft heart sounds. Early on you may hear a pericardial rub, but the accumulation of fluid soon separates his pericardial surfaces and stops the rub. He has the signs and symptoms of heart failure, and an increased area of cardiac dullness. The severity of the signs of cardiac tamponade are related more to the rate at which fluid accumulates in his pericardium than to the volume of fluid in it. The diagnosis may be obvious, or if fluid has accumulated slowly, it may be difficult.
There are problems: (1) Any cause of cardiac failure may have distended his neck veins. (2) Although pulsus paradoxus strongly suggests a pericardial effusion, not all patients show it. (3) The X-ray finding of a large globular heart can also be due to gross cardiac enlargement without there being any fluid in his pericardium.
The great danger in putting a needle into his pericardial cavity to drain it is that: (1) You can easily penetrate his right ventricle, cause bleeding, increase the fluid in his pericardial cavity, and kill him rapidly. (2) You may cause ventricular fibrillation with the tip of the needle. Even so, in spite of these dangers, not aspirating his pericardium may be more dangerous than aspirating it.
Fig. 6-2a ASPIRATING THE PERICARDIUM. Insert the needle in his epigastrium immediately to the left of his xiphisternum. Incline it 45[de] to the horizontal and 10[de] towards the left. In this way, if it does prick his heart it is more likely to meet his thicker left ventricle than his thinner right auricle.
PUS IN THE PERICARDIUM See also Section 65.9 for cardiac tamponade as the result of trauma.
X-RAYS. A very large globular heart, often with venous congestion. Depending on what is causing his pericarditis, you may see basal shadows in his lungs, or pneumonia obscuring his heart.
ECG. Tachycardia, usually sinus rhythm, a raised S-T segment (nonspecific), an inverted T wave (late, nonspecific), low voltage QRS complexes (highly suggestive), pulsus alternans (highly suggestive).
THE DIFFERENTIAL DIAGNOSIS of the causes of pericardial effusion which may lead to tamponade is as follows in probable order of frequency in most of the Third World:
Suggesting tuberculosis[md]a history of cough, bloody sputum, weight loss and malaise. Patients with AIDS and tuberculosis are particularly likely to develop tuberculous pericarditis and pleural effusions.
Suggesting viral myocarditis[md]an influenza-like illness with generalized muscle pains. Early, you may hear a pericardial friction rub.
Suggesting a pyogenic bacterial cause[md]some other site of infection, such as pneumonia, meningitis, or measles with secondary staphylococcal infection. Often, there is some obvious site of infection, but not always (primary pericarditis).
Other causes of pericardial effusion that might cause tamponade include: uraemia, malignant deposits (only if they bleed seriously), collagen diseases, and the rupture of an amoebic abscess into the pericardium (rare).
Here are some causes of a large heart without fluid in the pericardial cavity:
Suggesting rheumatic heart disease (common) [md]valvular lesions; these are usually easily diagnosed.
Suggesting cardiomyopathy [md]an enlarged heart clinically and radiologically. The cardiac outline may be globular and closely simulate fluid in the pericardium.
Suggesting endomyocardial fibrosis (EMF)[md]atrioventricular incompetence left and right is usual. Eosinophilia.
PREPARATION. Find two assistants, one to watch the patient's ECG, or his pulse, and ready to resuscitate him if necessary, and another to fetch anything more that might be needed for resuscitation. Have the full resuscitation equipment available: laryngoscope, tracheal tubes, a sucker, oxygen, and an anaesthetic machine or an Ambu bag. Do an ECG while you are aspirating, or failing this ask someone to feel his pulse continuously.
EQUIPMENT. A needle inside a plastic cannula (needle- inside-cannula', A 15.2), a 3-way tap (less satisfactorily a 2- way one), and a 20 or 50 ml syringe.
ASPIRATION. Insert the cannula in his epigastrium immediately to the left of his xiphisternum. Incline it 45[de] to the horizontal and 10[de] towards the left. In this way, if it does prick his heart, it is more likely to meet his thicker left ventricle than his thinner right auricle.
DRAINAGE. Incise his linea alba and proceed upwards in the extraperitoneal plane until you reach his pericardium. Cautiously incise this and insert a drain.
CAUTION ! If he deteriorates suddenly with a pulse which you cannot feel: (1) Immediately remove the cannula. (2) Start external cardiac massage (A 3.6). (3) While you stop external cardiac massage briefly, ask your assistant to intubate him. Continue to control his ventilation (A 13.1).
Peritonitis
The area of a patient's peritoneal cavity is three times that of his skin, and is a huge area to become infected. So it is not surprising that the mortality from peritonitis is about 10%, even in good units.
Bacteria, both aerobic and anaerobic, can reach the peritoneum from inside the gut (often), from an outside wound (occasionally), or from the blood-stream (rarely). Peritonitis can also follow a laparotomy if this is carelessly done. It is particularly likely to complicate: a strangulated obstruction of the gut (10.3), appendicitis (12.1), a penetrating abdominal injury (66.2), pelvic sepsis in a female (PID, 6.6), a septic Caesarean section (6.7), a perforated peptic ulcer (11.2) or typhoid ulcer (31.8), the breakdown of an anastomosis, especially of the unprepared large gut, where this has not been protected by a proximal colostomy, amoebic colitis (31.11), the torsion of an ovarian cyst (20.7), acute cholecystitis (13.3), pancreatitis (13.9), or the rupture of a liver abscess (31.12). Sometimes, there is no cause (primary' peritonitis). The frequency of these causes differs geographically, so find out what the common causes are in your area.
You can reduce the risk of death from peritonitis if you: (1) Operate early[md]before a patient is very ill. (2) Take the necessary precautions to minimize the infection of his peritoneal cavity, when you do any laparotomy. So handle his tissues gently, anastomose his gut carefully, pack away potentially infected areas, control bleeding meticulously, and give him perioperative antibiotics when these are indicated (2.9).
Peritonitis develops through several stages, which need different treatment:
(1) Disease in an organ before the peritoneum over it is infected. For example, a patient may have the symptoms of peptic ulceration, appendicitis, or of typhoid fever, but no involvement of his peritoneum. At this stage, you may be able to treat the underlying disease and prevent peritonitis.
(2) Localized peritonitis. With proper treatment this localized peritonitis may resolve, and he may recover. A mass may form, but the toxic features of an abscess will not have appeared.
(3) Abscess formation around the organ responsible for the infection. Pus forms, but this is sealed off from the rest of his peritoneal cavity by loops of gut which are stuck to one another by a fibrinous exudate. His abscess may resolve, as an appendix abscess often does, or it may spread, as is common with a perforated duodenal or typhoid ulcer. The mass is bigger than in stage (2) above, and he now has toxic symptoms.
(4) Spreading peritonitis which may become generalized. If spread is incomplete, multiple abscesses form in his peritoneal cavity, particularly in his pelvis and under his diaphragm. If he is unlucky, all his abdominal organs are bathed in pus. If you operate and wash the pus out of his peritoneal cavity, more abscesses may form postoperatively.
Gut which is surrounded by pus usually develops ileus (10.13). If his peritonitis becomes generalized, his abdomen becomes silent and distends, as his gut fills with fluid and dilates. This fluid, and that which is lost into his peritoneal cavity, depletes his circulation, so that his blood volume, his blood pressure, and his urine output all fall, and his pulse rate rises. As peritonitis advances, his peripheral circulation fails, and he may develop septic shock (53.4).
Take a careful history. Ask for the symptoms of any underlying disease which may have caused peritonitis, such as previous dyspepsia, or a fever that may be typhoid. In a woman, enquire for symptoms suggesting PID (6.6).
Ask about pain. The pain of peritonitis is constant and is made worse by deep breathing, by coughing, and by movement. A patient with peritonitis is weak and thirsty, anorexic and nauseated. He vomits, and may have diarrhoea or be constipated.
The signs of peritonitis vary with the state of the disease: (1) An early sign is the failure of his abdomen to move as he breathes. (2) His abdomen is tender. This tenderness is localized at first, then regional, and finally general. It is usually worse where the disease started. (3) Feel for muscle guarding, progressing to rigidity. If peritonitis is advanced, there is no need to test for rebound tenderness[md]it is painful and unhelpful. But if peritonitis is localized, rebound tenderness is a good indication as to which parts of his peritoneum are involved and which are not. (5) Listen for decreased or absent bowel sounds. (6) Look also for distension, eventually becoming tympanitic, as ileus develops. You will soon find that the intensity of a patient's signs is little guide to the nature of the fluid inside his abdomen. Gastric, bilious, and pancreatic fluids produce the most tenderness; pus, urine, and especially blood are more variable, and may cause almost none. Even faecal contamination produces no peritoneal signs at first.
As his peritonitis advances, he becomes febrile, apprehensive, dehydrated, and hypotensive. His pulse is fast, his breathing is shallow and his facies Hippocratic (pinched, drawn and grey). Finally, he dies in peripheral circulatory failure.
Often, all you will know before you operate is that he has peritonitis, without knowing why. Try to to establish how advanced it is. A laparotomy is usually mandatory but, as you will see below, there are some indications for non-operative treatment. If you are uncertain about these in a particular case, you would be wise to operate.
Aim to: (1) Resuscitate him by treating his dehydration for 2 or 3 hours before you operate (A 15.3). (2) Treat the cause of his peritonitis, for example by closing his perforated peptic ulcer (11.2), by removing his appendix (12.1), or by resecting and anastomosing his gangrenous small gut (9.3). (3) Remove the pus in his peritoneum (if his peritonitis is generalized), by thorough lavage, and then leave some tetracycline solution in his peritoneal cavity. This will reduce the danger of abscesses forming in his peritoneum later, and will increase his chances of recovery.
You are often in a serious dilemma, when you drain an abdomen for peritonitis. If the pus is thin and adhesions few and light, there is no problem. But if loops of his gut are firmly stuck together, how radical should you be? If you don't separate them enough, you will leave pockets of pus behind. If you separate too vigorously, you risk stripping off the muscle layer or opening his gut[md]a real disaster, because a fistula will probably follow (9.14). This is difficult surgery, and there is no easy answer; only experience will teach you. You feel you cannot win, yet you will have to! Sometimes you will fail, and have to reopen his abdomen.
About 5 days after the operation his tachycardia should subside, his temperature should settle and his bowel sounds should return. When this happens the volume of his nasogastric aspirate will decrease, his abdominal distension will go down and he will start to pass flatus. If he survives the first 10 days, he will probably live.
His postoperative course is likely to be stormy. It may be complicated by paralytic ileus (10.13), the formation of more abscesses (6.3), Gram-negative septicaemia (53.4), intestinal obstruction (10.13), or extreme nutritional deficiency (9.11). One of the greatest dangers is a faecal fistula (9.14). If you separate the adhesions between the loops of his gut too roughly, you may: (1) open it and have to repair it immediately, or (2) strip part or all of its muscular coat and weaken it, so that it breaks down later to form a fistula (9.14).
Fig. 6-3 THE PERITONEAL CAVITY. In generalized peritonitis this fills with pus. A, the posterior abdominal wall showing the lines of peritoneal reflection after removal of the liver, spleen, stomach, jejunum, ileum, and the transverse and sigmoid colons. Organs on the back of the abdominal wall are seen through the posterior parietal peritoneum. B, a longitudinal section of the abdomen.
1, the anterior superior subphrenic space. 2, the anterior inferior subphrenic space. 3, the lesser sac. 4, coils of jejunum. 5, the transverse colon. 6, the great omentum. 7, coils of ileum. 8, the bladder. 9, the rectum. 10, the mesenteric artery. 11, the duodenum. 12, the pancreas. 13, the liver. 14, the stomach. 15, the left triangular ligament of the liver. 16, the oesophagus. 17, the upper recess of the omental bursa (lesser sac). 18, the lienorenal ligament. 19, the root of the sigmoid colon. 20, the root of the mesentery. 21, the cut edge of the lesser omentum. 22, the ascending colon. 23, the descending colon. 24, the duodenum.
GENERALIZED PERITONITIS This extends the general method for a laparotomy in Section 9.2. For tuberculous peritonitis see Section 29.5.
X-RAYS. Take an erect and a supine film. Look for: (1) free air under the patient's diaphragm. (2) Gas and fluid filled adjacent loops which appear to be separated, due to the exudate between them. (3) Distended loops of gut; to recognize the different levels of the gut radiologically, see Section 10.4 and Figs. 10-6 and 10-7.
BLOOD COUNT. He is likely to have a leucocytosis of more than 15,000 [gm]l. If his haematocrit is over 50% he has lost much extracellular fluid and needs Ringer's lactate.
ASPIRATION. If you suspect acute pancreatitis (13.9), confirm it by aspirating his peritoneal cavity with a needle, and if this is negative do a peritoneal lavage (66.1).
RESUSCITATION. The need for this varies:
If his peritonitis is early, and his general signs are minimal, he does not need resuscitation.
If his pulse is rapid and his blood pressure low, delay operation for a few hours (never more than 6) while you resuscitate him (A 15.3). Give him intravenous Ringer's lactate or saline (an adult may need several litres). If possible, measure his CVP and keep it at 6[nd]8 cm of water (A 19.2). Monitor his hourly urine output (A 15.5) and keep a fluid balance chart. Catheterise his bladder. Operate as soon as his pulse rate falls, his blood pressure rises, and his peripheral circulation improves.
If signs of peripheral circulatory failure do not respond to generous resuscitation, he will probably die, whatever you do. You may be wise not to operate.
If he is confused, severely hypotensive, and hyperventilating, with a fast pulse, and warm pink extremities, or cold clammy ones, he is in septic shock. Treat him as in Section 53.4. If you can, drain the septic focus. Timing is important: he must be fit enough to withstand the operation, so overcome shock, and then do the simplest possible operation.
THE NON-OPERATIVE TREATMENT [s7]OF PERITONITIS INDICATIONS. You have got to be very sure about these: (1) Acute pancreatitis. (2) Some cases of typhoid peritonitis (31.8). (3) Peritonitis which is mainly pelvic. You can feel an inflammatory mass vaginally or rectally. Drain this pus vaginally after confirming it by aspiration (16.6), or rectally in a male (uncommon). (4) Pus which is mainly under the diaphragm (6.4). (5) Peritonitis which has been confirmed by aspiration (66.1), but the patient is too ill to withstand laparotomy. Delay operation until he has improved.
METHOD. Start antibiotics and nasogastric aspiration as described below and give him intravenous fluids (A 15.3, 15.5). Give him nothing by mouth. Be sure to correct potassium deficiency (A 15.3). Continue nasogastric suction and intravenous fluids until he shows signs of recovery (his bowel sounds return, there is less aspirate, and he passes flatus).
LAPAROTOMY [s7]FOR PERITONITIS NASOGASTRIC ASPIRATION. Insert a nasogastric tube (4.9).
ANTIBIOTICS are unnecessary if he has: (1) Acute pancreatitis. (2) A perforated gastric or duodenal ulcer (unless you see him late when peritonitis has developed, 11.1). (3) Appendicitis causing only localized peritoneal infection. Otherwise, give them.
If he has generalized peritonitis, take blood cultures (if possible) before you give him antibiotics. You have a good chance of isolating the organisms responsible. When the sensitivity tests come back, adjust his antibiotics accordingly. Meanwhile, give him the perioperative antibiotics, as in Section 2.9. Chloramphenicol or a cephalosporin, and metronidazole are likely to be the practical ones. Give him chloramphenicol 6-hourly, at first intravenously, later orally; and metronidazole 8-hourly, the first two doses intravenously if possible and rectally later. For tetracycline instillation, see below.
CAUTION ! If he is to be given a relaxant, don't give him an aminoglycoside antibiotic unless you have an experienced anaesthetist[md]it may prolong the paralysis (A.14.2 to 14.4). These include gentamicin (especially), kanamycin, streptomycin, and amikacin.
EQUIPMENT. A general set (4.12). Several litres of warm saline or Ringer's lactate. To warm them, see below.
INCISION. As soon as he is draped, and anaesthetized, and his abdomen is relaxed, palpate it (10.1). Unless you have good indication for making another incision, make a median or a right paramedian one, centred on his umbilicus.
For a list of some of the things you might find on opening his abdomen, see Section 10.2. We assume here that he has localized or generalized peritonitis, with pus and fibrinous exudate everywhere. First, take a specimen for culture and sensitivity.
Break down adhesions with the greatest possible care. Only break down light ones with your fingers. If they are dense, define them carefullly, and cut them with scissors, or, better (if you are experienced) with a fine scalpel (10-11). If you are rough, you increase the chances of a faecal fistula.
If peritonitis is widespread, search systematically until you can find its cause. Be guided as to where it might be by: (1) His history. (2) The nature of the exudate. (3) The place where pus and exudate are most intense. (4) The density of the adhesions; the densest ones may indicate the origin of the infection.
CAUTION ! (1) Suck out all free pus before you start. (2) You must have good exposure[md]see Section 9.2. (3) If you find localized pus, try to minimize its spread around his peritoneum! (4) You face the dilemma described above[md]when to divide adhesions and when not to.
MANAGING THE UNDERLYING CAUSE. First, you will have to find it, and this may not be easy. Look for appendicitis (12.1), PID (6.6), a perforated peptic ulcer (11.2), strangulation obstruction (10.3), and signs of typhoid fever (multiple lesions in his distal small gut, 31.8). In most cases of typhoid, you will probably have decided not to operate. If your search produces something that you can do easily, without breaking down too many protective adhesions, such as removing a gangrenous appendix when the tissues are not too friable, or an infected ovarian cyst, do it. If you cannot find a cause after a full laparotomy, lavage his peritoneum and instil tetracycline. Play safe: he is desperately sick, and you must not risk complications.
If he has a hole in his large gut, repair it and make a proximal defunctioning colostomy (9.5).
If he has a hole in his rectum, do a proximal colostomy or a Hartmann's procedure (9.6).
If a dilated loop of gut disappears' into an inflammatory mass that might be tuberculosis, or a sealed-off perforation, don't try to dissect out the mass. Bypass it (29.5). This will keep risk to a minimum, relieve incipient obstruction, and allow the inflammation to subside. If you can easily biopsy the mass, do so. Usually, plan to re-operate and resect the lesion 3 to 6 months later.
If he has a perforated peptic or typhoid ulcer, oversew it and apply an omental graft (11.2).
LAVAGE. If he has generalized peritonitis, lavage his peritoneal cavity with saline. If his peritonitis is localized this may only spread the infection, so don't lavage. Sometimes, you can safely wash out only the pelvis.
Tip in several litres of warm saline or Ringer's lactate with 1 g of tetracycline (oxytetracycline may be cheaper) to the litre, slosh it around with your hand, and suck it out until the fluid which returns is clear. You may need 8 to 10 litres. Usually 3 or 4 are enough. Wash out his upper abdomen, his paracolic gutters, his infracolic area and his pelvis. Mop his peritoneum dry. Finally, leave a litre of warm tetracycline solution in his peritoneal cavity.
To warm the saline, put the bag or bottle in a basin of hot water and warm it to blood heat, feeling its temperature with your hand. If you have no saline you will have to use water, but the last instillation containing the antibiotic should be saline, or some other isosmotic fluid.
CLOSURE. Close his abdomen with interrupted through-and- through sutures of stout monofilament nylon or steel deep to the skin (9.8). Leave his skin unsutured for delayed closure.
DRAINS. Generalized peritonitis and multiple intra- abdominal abscesses cannot be drained adequately, because the area to be drained is too large: so wash out the pus, instil tetracycline, and don't insert drains.
It may be appropriate to drain a localized abscess[md]a pelvic abcess (vaginally in a female or rectally in a male), or a single intra-abdominal abscess. Abdominally, use wide bore tube drains; rectally, use corrugated rubber (4.10).
CAUTION ! If you instil tetracycline, don't insert drains, or it will all flow out!
POSTOPERATIVE CARE [s7]FOR PERITONITIS FLUID BALANCE. Continue to suck and drip' him (9.9, A 15.5), and keep an accurate fluid balance chart. The common error is not to give him enough fluid.
Nasogastric suction. If he has had generalized peritonitis, he is sure to get ileus; suction will reduce his distension. You may suck out 2 to 6 litres of fluid a day. Replace it with 0.9% saline or Ringer's lactate in addition to his standard requirements (A 15.5).
Intravenous fluids. Manage his fluid balance as in Section A 15.5. For maintenance an adult needs at least a litre of 0.9% saline, or Ringer's lactate, and 2 litres of 5% dextrose in 24 hours. Be sure to monitor his urine output (if possible 2- hourly for the first 48 hours). After the initial period of up to 48 hours, when you expect his urine output to fall, keep his urine output above 1 ml/kg/hour. Replace all losses as appropriate (A 15.5). If his initial resuscitation was inadequate, he may still have a deficit to make up.
Potassium supplements. Don't forget these (A 15.5), especially if there is a large volume of gastric aspirate. Start them when his postoperative diuresis begins.
He may be acidotic. There are several ways you can correct this. You can: (1) Give him 200 ml of 8.4% sodium bicarbonate (200 mmol). Or, give him 500 ml of 4.2% sodium bicarbonate (250 mmol). (2) Give him a litre of 1/6 molar lactate. (3) Give him adequate intravenous fluids and let his kidneys correct his acidosis. If his condition is poor, use (1) or (2), and repeat them daily.
Blood. If he bled during the operation, and this loss was not replaced, replace it now.
POSTOPERATIVE ANTIBIOTICS. If he had generalized peritonitis, continue the same antibiotics you gave him preoperatively for 5 to 7 days. Be guided by his clinical response, rather than by the sensitivities reported by the laboratory. If he has not improved after 3 days change them.
OTHER MEASURES. Examine him carefully each day for complications. Watch his temperature chart, his general state of alertness, his abdominal girth, his bowel sounds, and the volume of his gastric aspirate.
DIFFICULTIES [s7]WITH PERITONITIS These are many, and include septic shock (53.4), which can develop postoperatively.
If you CANNOT FIND A CAUSE FOR PERITONITIS, remember that PRIMARY PERITONITIS without any obvious cause does exist and is not uncommon in Africa. Bacteria may have arrived in the bloodstream as part of a septicaemic or pyaemic process. It is a diagnosis of exclusion, so make sure there is no perforation in any part of the gut, no PID or external injury etc.
If he DOES NOT IMPROVE, he may have residual sepsis and need a further laparotomy.
If his ABDOMEN DISTENDS and the volume of his gastric aspirate remains high (or he vomits), either the normal short period of ileus is continuing, or his gut is obstructing, see Section 10.13.
If his WOUND BECOMES INFECTED and breaks down, see Section 9.12. This is rare if you use non-absorbable sutures and close all the layers of his abdomen, except his skin (which should be left open for delayed closure), as a single layer. If you use absorbable sutures and close his abdomen layer by layer, wound breakdown is more likely.
If FEVER CONTINUES he may have a postoperative urinary or chest infection, or any of the abscesses in Section 6.3.
If he has DIARRHOEA, especially with the passage of mucus, suspect a pelvic abscess (6.5).
If FAECES START TO DISCHARGE FROM THE WOUND or a drain, he has a faecal fistula (not uncommon), so see Section 9.14. This is usually due to stripping some of the muscular coat of his gut, as you separate adhesions, and weakening it so that it breaks down later. If the fistula persists, it may produce disastrous fluid losses and severe wasting. Finding and closing it will be very difficult, so manage him non-operatively at first. If his gut is not obstructed distally, his fistula may close spontaneously. If it fails to close, refer him if you can. Try to avoid a fistula by only dividing light adhesions with your fingers. Define all other adhesions clearly, and divide them with scissors or a scalpel. Don't use diathermy close to his gut.
Abscesses in the peritoneal cavity
Localized abscesses in a patient's peritoneal cavity can be the result of: (1) Generalized peritonitis[md]they are one of its major complications. (2) Some primary focus of infection, such as appendicitis or salpingitis (PID, 6.6). (3) An abdominal injury in which his gut was perforated or devitalized. (4) Any laparotomy.
If abscesses are going to occur after a laparotomy, a patient's temperature does not fall, or it falls and then rises in a characteristic spiky pattern (Fig. 6-6) which shows that there is pus somewhere inside him. He is not well and does not eat, he loses weight, and his white count is raised. If loops of his gut pass through the abscess, they may become obstructed, acutely or subacutely (10.12).
Pus can gather: (1) Under his diaphragm or his liver (6.4). (2) Between loops of his small gut in the folds of his mesentery. (3) In his pelvis. If you are wondering where it has got to, remember that pus usually follows recognized paths. For example, in escaping from the appendix, it usually falls into the pelvis or tracks up his right paracolic gutter to his right subphrenic space, so that you are unlikely to find it on the left of his abdomen.
Provided you remember to examine a patient rectally (or vaginally), diagnosing pus in the pelvis should be easy. Diagnosing it under his diaphragm is much more difficult.
If you catch a localized infection early enough, antibiotics alone may possibly cure him; but once pus has formed they will not make it disappear, although they may limit its spread. So be sure to drain residual abscesses on the indications given below.
Fig. 6-4 ABSCESSES IN THE PERITONEAL CAVITY. A, the common sites. B, a pelvic abscess pointing into the rectum. C, a pelvic abscess pointing into the vagina.
1, between a patient's diaphragm and his liver. 2, under his liver. 3 and 4 in his right and left paracolic gutters. 5, among coils of his gut. 6, around his appendix. 7, in his pelvis.
ABSCESSES [s7]IN THE ABDOMEN If you suspect that a patient has an abdominal abscess, record his temperature 4-hourly, or hourly if he is very ill, especially if he is a young child. Examine him carefully at least once a day. Each time, feel for an abdominal mass, feel under his rib margins anteriorly and posteriorly, do a rectal examination, and in a woman a pelvic examination. The patient's temperature chart and his clinical signs will be of the greatest help, but you may find it helpful to do the following examinations each day: an abdominal X-ray (erect and supine), a chest X-ray, a white count, and blood cultures.
If you feel a mass, mark it out on his abdominal wall. Each day, feel if it has become larger or smaller. If it becomes larger, operate. It may become adherent to his abdominal wall, so you can open it without opening the rest of his peritoneal cavity.
On the appropriate indications, drain pus from his wound (9.12), from between the loops of his gut (see below), from under his diaphragm (6.4), and from his pelvis (6.5). Drainage is paricularly urgent if his general condition is deteriorating, or if he has complete intestinal obstruction which has not responded in 24 hours (10.12).
CAUTION! Don't make a small abdominal incision, his gut will be in less danger if you make a large one.
If you can feel a pelvic abscess vaginally, aspirate it to confirm the presence of pus, then drain it vaginally (6.5). Only drain an abscess abdominally, if you cannot drain it vaginally.
ABSCESSES BETWEEN LOOPS OF GUT. If the swelling is to one side of a patient's abdomen, incise its lateral side. Open the layers of his abdominal wall, then explore his abdomen with your finger until you find pus. If infection is localized, insert a drain.
POSTOPERATIVELY, after you have drained any kind of abscess, watch him carefully, he may have more. Don't neglect his fluid balance or his nutritional state. He will be wasting severely, so do your best to increase his protein and energy intake (9.10). Ideally, he needs feeding parenterally, which is likely to be impossible. Don't try forced feeding, because he will not want to eat. If he will tolerate cautious feeding through a nasogastric tube, he may benefit considerably.
Subphrenic abscess
Pus under a patient's diaphragm has usually spread there from somewhere else in his abdomen. A subphrenic abscess may be secondary to: (1) Peritonitis, either local or general, following a perforated peptic (11.2) or a typhoid ulcer (31.8), or appendicitis (12.1), or PID (6.6) or infection following Caesarean section (18.11). (2) An injury which has ruptured a hollow viscus and contaminated his peritoneal cavity (66.2). (3) A laparotomy during which his peritoneal cavity was contaminated (9.2). (4) A ruptured amoebic liver abscess (31.12).
Suspect that a patient has a subphrenic abscess if he deteriorates, or recovers and then deteriorates, between the 14th and the 21st day after a laparotomy, with a low, slowly increasing, swinging fever, sweating, and a tachycardia. This, and a leucocytosis, show that he has pus somewhere', which is making him anorexic, wasted, and ultimately cachectic. If he has no sign of a wound infection, a rectal examination is negative, and his abdomen is soft and relaxed, the pus is probably under his diaphragm.
The pus might be between his diaphragm and his liver, in (1) his right or (2) his left subphrenic space, or under his liver in (3) his right or (4) his left subhepatic space in his lesser sac. He may have pus in more than one of these spaces.
Explore him on the suspicion that he might have a subphrenic abscess. Exploration is not a major operation; the difficulty is knowing where to explore, so refer him if you can. If you cannot refer him, explore him yourself. If you fail to find pus, you have done him no harm; missing a subphrenic abscess is far worse. If it is anterior, you can drain it by going under his costal margin anteriorly. If it is posterior, you can go through the bed of his 12th rib posteriorly.
KIMANI (15 years) was admitted with abdominal pain and vomiting of sudden onset, about 4 hours previously. He had shoulder-tip pain, but he also said he had pain when he put his tongue out, so it was first thought that he might be hysterical. He had no abdominal signs, so he was admitted for observation. The following day his abdomen started to distend, and aspiration of his peritoneal cavity withdrew greenish fluid. A laparotomy was done, and an ulcer on the greater curve of his stomach was found and repaired. Initially he recovered well, but as he was about to go home 10 days later, he was not well, he ran a fever, he looked toxic, and there was tenderness and induration on the right side of his upper abdomen. He was suspected of having a subphrenic abscess, his abdomen was reopened through his paramedian incision, and a large quantity of foul-smelling pus was evacuated from under the right side of his liver. Several drains were inserted. The drug vote was almost finished and the hospital could not afford intravenous metronidazole, so he was given intravenous chloramphenicol, and metronidazole by mouth, after which he eventually recovered. LESSONS (1) If you are not certain that a patient is hysterical it always pays to observe him. (2) Beware of the latent interval' 3 to 6 hours after a perforation, when there may be few abdominal signs. (3) You may be able to drain a subphrenic abscess through the original laparotomy incision, but the incisions described below may be better. (4) When a peptic ulcer causes general peritonitis, a thorough lavage of the peritoneal cavity is as important as the repair. Fig 6-5 SUBPHRENIC ABSCESSES. A, the spaces where pus can collect under a patient's diaphragm. B, make a subcostal incision when you approach a subphrenic abscess anteriorly. C, exploring the right posterior subphrenic abscess. D, exploring the left posterior space. E, exploring the posterior spaces through the ribs.
1, the right anterior subphrenic space. 2, the left anterior subphrenic space, 3, the right subhepatic space. 4, the left subhepatic space (the lesser sac). 5, the right posterior subphrenic space. 6, the left posterior subphrenic space.
SUBPHRENIC ABSCESS This follows from Section 6.3 on abdominal abscesses.
SIGNS AND SYMPTOMS. Thoracic signs are more useful than abdominal ones. Ask or look for: (1) Cough. (2) Shoulder-tip pain on the affected side. (3) An increased respiratory rate, with shallow or grunting respiration. (4) Diminished or absent breath sounds. (5) Dullness to percussion. (6) Dull pain. (7) Hiccup (rare). (8) Tenderness over the 8th to 11th ribs. A subhepatic abscess may cause tenderness under the costal margin anteriorly. A subphrenic abscess, pyelonephritis, a pyonephros or a perinephric abscess can all cause similar tenderness posteriorly. (9) If the patient is thin and the pus is superficial, you may feel a tender indurated mass under his costal margin in front (right subphrenic space), in his right flank (right subhepatic space), or posteriorly.
X-RAYS are essential. Screening is the most important investigation and the cheapest. Look for: (1) The failure of one side of his diaphragm to move. This is a sign of infection, but not necessarily of an abscess. (2) Give him a little contrast medium, and look for downward and forward displacement of his stomach and spleen.
Also take a PA and a lateral view. Look for: (1) a raised diaphragm, (2) a fuzzy upper border to his diaphragm, (3) fluid in his costophrenic angle, (4) collapse or consolidation at one lung base, (5) a fluid level (rare). (6) You may also see gas in his subphrenic space. This can be the residue from a laparotomy, or it can be due to a perforation of his gut, or to an anaerobic infection.
If his first X-ray examination is negative or equivocal, repeat it a few days later.
CAUTION ! (1) His white count is usually raised but may be normal. (2) 10% of patients have no fever. (2) Don't try to diagnose subphrenic abscesses by aspiration[md]this is dangerous and misleading.
THE DIFFERENTIAL DIAGNOSIS includes a liver abscess (31.12), an empyema (6.1), and pulmonary collapse (9.11).
THE MANAGEMENT [s7]OF SUBPHRENIC ABSCESSES WHICH APPROACH? If you suspect a subphrenic abscess, and a patient's general state does not improve, and his fever does not settle, he needs exploring. Avoid antibiotics which may mask his symptoms.
If he has a swelling, or oedema, or redness or tenderness just below his ribs or in his loin, make the incision there.
If his abscess follows appendicitis, a perforated duodenal ulcer, or cholecystitis, it will probably be on the right. If a high gastric ulcer has perforated, it is more likely to be on the left. If an ulcer in the posterior wall of his stomach has perforated, there will be pus in his lesser sac.
If you don't know which side it is on, there is about a 75% chance that it will be on the right, probably anterior. Approach it anteriorly, if possible through the old laparotomy wound, unless there are very clear signs that it is posterior. If one route fails try another. You cannot reach the posterior surface of his liver through an anterior incision, or vice versa, but if pus extends all the way from front to back, one incision will be enough.
Alternatively, decide if the pus is on the right or left, and then explore all subphrenic abscesses from in front. If you don't find pus, explore posteriorly.
ANAESTHESIA. Take him to the theatre. If he is a poor anaesthetic risk, block his lower 6 intercostal nerves (A 6.7). If he is a better risk, you can give him a general anaesthetic and intubate him.
ANTERIOR APPROACH. Make an incision which is large enough to take your hand. Depending on the signs, make it a finger's breadth below and parallel to his right (usually) or his left costal margin. Cut from the middle of his rectus muscle laterally, as in B, Fig. 6-5. Cut the muscle fibres in the line of the incision. Often, you can open the abscess cavity without entering his general peritoneal cavity, so try to keep outside it until you have found the abscess. His extraperitoneal tissue will probably be oedematous. Push your index finger upwards through it, peeling the peritoneum off his diaphragm as you do so. Sweep your finger under his liver from one side to the other to explore his subhepatic space. If you don't find pus there, sweep it round the lateral edge of his liver, and explore his subphrenic space between his diaphragm and his liver.
Somewhere you will feel an indurated abscess cavity. If you have opened his general peritoneal cavity pack off his gut, and have a sucker ready before you push your finger through it, in case pus squirts out. Explore it with your hand, break down any loculi, and send pus for culture. Insert a drain as described below.
If his liver is not adherent to his diaphragm, there may still be pus posteriorly, pushing his liver forwards.
CAUTION ! (1) Try not to go above his diaphragm. This is more likely to happen with an anterior approach. If you enter his pleura, suture his diaphragm with 1' multifilament or monofilament sutures and insert an under water seal drain (9.2D, 65-6) before you approach the abscess. (2) Be sure that he has only one abscess.
Alternatively, cut everything except his parietal peritoneum in the line of the incision. Burrow upwards with your finger between his peritoneum and his diaphragm, until you feel the induration of the abscess. Peel his peritoneum off his diaphragm as you do so.
Fig. 6-6 THE POSTERIOR APPROACH TO A SUBPHRENIC ABSCESS. A, a hectic fever subsided as the patient's abscess was drained. B, his 12th rib has been excised and an incision is about to be made in its bed. C, the bed of his 12th rib has been divided, showing his liver and the fat round his kidney. After Ochsner and Graves. POSTERIOR APPROACH. Lay him on his sound side with his lumbar region slightly elevated by breaking up the table or placing pillows under his other side. Make an incision which is big enough to take your hand over his 12th rib posteriorly (E, in Fig. 6-5). Remove the distal 2/3 of his 12th rib; divide it at its angle. Cut through the periosteum, reflect this from the whole circumference of the bone with Faraboef's rougine, as you would when you drain an empyema (6.1).
CAUTION ! Take great care not to damage his diaphragm.
Incise the inner aspect of the periosteum horizontally. Push your finger upwards and forwards above his renal fascia to enter the abscess (C, or D, in Fig. 6-5). Occasionally, you may need to tie his intercostal vessels.
POSTOPERATIVELY (both routes). At the most dependent part of the abscess, insert one or even two 1.5 cm plastic or rubber drainage tubes with several side holes, or a sump drain. Bring the drain out through a stab wound in his flank. Stitch it to his skin. Close his wound as usual, but leave his skin unsutured. If you have left a large space under his diaphragm, connect the drain to an underwater seal, to encourage it to close.
As soon as the discharge is reduced to about 20 ml of pus a day, shorten the drain progressively during a few days and then remove it. Sinograms are unlikely to be helpful.
If there is any rise in his pulse or temperature, or localized pain, suspect that his abscess is not settling. Be sure to leave the drain in. Antibiotics are less important than adequate drainage.
DIFFICULTIES [s7]WITH SUBPHRENIC ABSCESSES If the fluid you aspirated from his chest was a STRAW-COLOURED EFFUSION, and he is not very toxic, X-ray him again in a few days to see what has happened. Is it clearing? Is his diaphragm still raised? If you aspirated frank pus, drain it a day or two later by inserting a chest tube connected to an underwater seal (65.2).
If he is so toxic, weak, and wasted from his subphrenic abscess that SURGERY MIGHT SEEM TO BE TOO MUCH FOR HIM, don't hesitate to explore his abscess. If necessary, drain pus from his pleural space also, it is his only hope.
If you DAMAGE HIS PLEURA ACCIDENTALLY, insert an underwater seal drain.
If pus from below the diaphragm RUPTURES INTO A BRONCHUS, he may drown in a spasm of coughing. The pus is more likely to have spread from an amoebic abscess in his liver than from a subphrenic one.
Pelvic abscesses
Pus in the pelvis is nearly as dangerous and difficult to manage as pus under the diaphragm. You will see several kinds of pelvic abscess which need managing in different ways:
(1) Following infection of the female genital tract which can be any of the varieties of pelvic inflammatory disease (PID) described in the next section. Those following a septic abortion or puerperal sepsis may be caused by anaerobes, and so are particularly serious and likely to spread. The patient may be very ill; you may have difficulty finding pus, and knowing when and how to drain it. The danger is that pus may build up as a mass above her pelvis, and spread upwards into her peritoneal cavity, perhaps fatally, instead of discharging spontaneously and harmlessly into her rectum. Drain this type of pelvic abscess early, as soon as pus has formed, and don't sit on it'.
(2) Following appendicitis. You can treat pelvic abscesses of this type non-operatively with much more safety, as in Section 12.2.
(3) Following generalized peritonitis, such as that caused by a perforated peptic (11.2), or a typhoid ulcer (31.8). If a patient makes good progress from the disease which caused his abscess, the pus in his pelvis is unlikely to kill him.
A pelvic abscess can grow quite large without making a patient very ill, or causing very obvious signs, so that, unless you do frequent rectal or vaginal examinations, you can easily miss one. You will need experience and a sensitive finger. One danger is that a pelvic abscess may obstruct the gut (10.13). Drain a man's abscess rectally, and, if possible, a woman's vaginally. This is easier then doing a laparotomy. If coils of gut lie between the pus and her posterior fornix, it will be more difficult to diagnose, and you will have to drain it suprapubically. Sometimes, an abscess drains into the rectum spontaneously, but incising it speeds recovery.
Fig. 6-7 DRAINING PELVIC PUS. A, if you drain pus vaginally, do so with particular care, or you may penetrate coils of gut. B, occasionally, pus presents above the inguinal ligament and has to be drained like this.
PELVIC ABSCESSES This follows from Section 6.3 on abdominal abscesses.
DIAGNOSIS. PID is the commonest cause. Watch also for the formation of a pelvic abscess when a patient is recovering from appendicitis (12.1), peritonitis, or an infected Caesarean section (6.8); watch for fever and the passage of frequent stools (diarrhoea'), with tenesmus and mucus. Feel for: (1) A boggy, tender mass above a man's prostate filling his rectovesical pouch, or a soft bulging swelling in a woman's pouch of Douglas. Sometimes, the mass is almost visible at her vulva. You will not find fluctuation. (2) Tenderness and occasionally an ill-defined mass suprapubically. If you suspect a pelvic abscess in a woman, put one finger into her rectum and another into her vagina. Normally, they should almost touch. If she has an abscess, you will feel it between your fingers.
You can confuse a pelvic abscess bulging into the pouch of Douglas with: (1) A chronic ectopic pregnancy (haematocoele, 16.7). (2) An ovarian cyst (20.7). Some suspected cysts turn out to be post inflammatory collections of fluid (post-inflammatory pelvic pseudocysts).
If there are no signs that the infection is spreading upwards into the peritoneal cavity, operation is not urgent. Give antibiotics as for peritonitis (2.9, 6.2). Carefully monitor the patient's temperature and the mass, and drain the abscess as soon as it is ripe. If you doubt whether it is ripe' for drainage or not, wait.
CAUTION ! An abscess which is enlarging suprapubically needs draining urgently.
VAGINAL DRAINAGE [s7]OF A PELVIC ABSCESS (posterior colpotomy) INDICATIONS. A pelvic abscess which extends into the pouch of Douglas.
ANAESTHESIA. (1) General anaesthesia. (2) Ketamine (A 8.1).
PREPARATION. Put her into the lithotomy position and catheterize her bladder. Do a vaginal examination to confirm the diagnosis.
INCISION. Expose the vaginal wall of her posterior fornix (which should be bulging) with a short broad speculum. Ask an assistant to depress her vaginal wall with a Sims speculum, while you raise the posterior lip of her cervix with a vulsellum. Push a large needle into the swelling in the midline and aspirate:
If you aspirate pus, this confirms a pelvic abscess, so proceed as below.
If you aspirate aspirate a pale yellow fluid, you are probably draining a post-inflammatory pseudocyst, so also proceed as below.
If you aspirate blood, either you have punctured a blood vessel (which should not happen if the needle is in the midline; the blood will clot), or she has a haematocoele due to a chronic ectopic pregnancy (if so the blood will not clot), for which you must do a laparotomy (16.7).
If you find an abscess or a post-inflammatory cyst, make a 2[nd]3 cm transverse incision in her vaginal wall in the place where you found pus. Push in a haemostat; pus or fluid should flow. Open the forceps and pull them back to enlarge the opening. Explore the abscess with your finger; feel for loculations in the abscess cavity and gently open them. Insert a large drain and suture it to her perineum or labia. Leave it in for a few days and continue antibiotics. Pus may discharge for up to 2 weeks.
CAUTION ! (1) If pus is pointing laterally, drain it as close to the midline as you can, to avoid injuring her ureters (20-9). (2) Don't push too deeply into the abscess with the haemostat, or its roof may give way and spread the pus into her peritoneal cavity; or you may damage a loop of gut. Be safe, and gently insert your finger through an adequate incision. (2) The effect should be spectacular, and she should improve markedly in a few days. If she does not improve, she has more pus somewhere, and probably needs a laparotomy.
RECTAL DRAINAGE [s7]OF A PELVIC ABSCESS Take the patient to the theatre and anaesthetize him. Put him into the lithotomy position. While his abdomen is relaxed, palpate it gently. Then examine him bimanually with one finger in his rectum, and your other hand on his abdomen. If you can ballot the mass, needling is unnecessary [md]drain it immediately.
To needle it, take a three-ringed 10 ml syringe, as used for injecting piles, and fix a 1 mm needle to it. Place the tip of your gloved right index finger over the place in the anterior wall of his rectum where you feel pus. Slide the point of the needle up alongside your finger, then push it through the wall of his rectum for about 2 cm. Aspirate. If no pus comes, inject a few ml of saline, and aspirate again. The needle may be blocked.
If you aspirate pus, or are sure that his abscess is ripe, drain it. Either push the tip of your index finger into it[md]pus will burst out. Or take a long curved haemostat, and with your index finger again acting as a guide, push its tip through the anterior wall of his rectum into the abscess. Enlarge the hole by opening and closing the jaws a few times.
CAUTION ! (1) Opening an abscess before it is properly formed is useless and dangerous. (2) Don't use any sharp instrument to penetrate the rectal wall, it may bleed seriously.
SUPRAPUBIC DRAINAGE [s7]OF A PELVIC ABSCESS This is sometimes needed in women (it is almost never necessary in men), particularly after an abortion or a Caesarean section when you can feel a mass suprapubically but not vaginally. Fortunately, you can usually drain an abscess from below, which is easier and safer. Rarely, if more pus collects after vaginal drainage, you may need to drain it suprapubically.
If she is distended and tender, and there is induration behind and above her pubis, especially if she is also severely toxaemic, drain the pus suprapubically.
Catheterize her bladder to make sure it is empty. Make a 10 cm midline incision immediately above her pubis. Incise her linea alba and her peritoneum. If you enter her general abdominal cavity (which you can usually avoid doing), inspect it first, then pack off her upper abdomen with some large moist abdominal packs. Gently feel for the abscess. Look for pus, for loops of gut stuck down in her pelvis, and for oedematous or congested tissues. Insert a self-retaining retractor. Use a swab on a stick' to gently mobilize adherent loops of gut, until you have found the pus.
CAUTION ! (1) Don't lower her head to improve exposure; this may spread the infection. (2) Keep manipulation to a minimum. (3) When you have found pus, do nothing more than is necessary to ensure adequate drainage. Don't break down the outer walls of the abscess cavity, but do break down any loculi. Distinguishing between them may be difficult.
Culture the pus and insert a drain. Remove all the packs; suture her abdominal muscles securely, but do not close her skin immediately (9.8).
DIFFICULTIES [s7]WITH PELVIC ABSCESSES If a patient has COLICKY PAIN, VOMITING, AND ABDOMINAL DISTENSION, his small or his large gut is obstructed. Try to treat him non-operatively, with nasogastric suction and intravenous fluids (9.9, 10.13, A 15.5). Draining his abscess will usually cure the obstruction. If it does not, you may have to relieve it operatively (10.12).
Fig. 6-8 CAUTION! PID AND HIV. The organisms responsible for PID may be: 1, sexually transmitted (gonococci, mycoplasma, or chlamydia). 2, The normal flora of the patient's gut and vagina (coliforms, anaerobes, and, rarely, actinomyces). Both partners are potentially in danger of HIV, see Chapter 28a. With the kind permission of the Daily Telegraph.
Infection of the female genital tract; pelvic inflammatory disease[md]PID
PID will probably be the commonest gynaecological disease you will see, and may account for a third your admissions. If it is common in your area, half the women of childbearing age who present with acute abdominal pain may have it. You may admit two or three of them every week, and treat ten times as many as outpatients. The numbers of these patients, their frequent long stay in hospital, their mortality, the surgery they need, and the complications that follow make PID a major public health problem.
Infection elsewhere in the abdominal cavity usually originates in the gut, but infection in a woman's pelvic cavity usually starts in her genital tract. With the rare exception of tuberculosis, it always ascends from her vagina and cervix. PID is thus only a disease of women.
Infection follows two routes:
(A) Infection may spread through her tubes to cause: (1) In her cervix, cervicitis. (2) In her endometrium, endometritis. (3) In her fallopian tubes, salpingitis or pyo- or hydro-salpinx. (4) In her tubes and ovaries, salpingo-oophoritis (acute, subacute, or chronic) or a tubo-ovarian abscess. (4) In her pelvic cavity, pelvic peritonitis or a pelvic abscess. (5) In the rest of her peritoneal cavity, generalized peritonitis or peritoneal abscesses.
(B) Infection may also spread through her uterine wall into her broad ligaments to cause cause pelvic cellulitis (parametritis), a broad ligament abscess, or septic thrombophlebitis of her ovarian or her uterine veins. This is serious and causes septicaemia with few local signs.
Both subacute and chronic PID may cause an inflammatory mass containing her inflamed tubes, her ovaries, her uterus, her omentum, and loops of her gut. Between all these there are collections of pus, and in chronic cases fluid-filled pseudocysts. PID is always bilateral, although it may be dominant on one side.
Although they are similar, it is convenient to discuss: (1) PID unrelated to pregnancy, that is PID which does not obviously follow abortion or delivery. Because this kind of PID typically follows a period, it is sometimes called postmenstrual PID'. For want of a better name, this is what we will call it here. It is one of the most serious effects of the three sexually transmitted organisms discussed below (gonococci, chlamydia, and mycoplasma). When PID' is referred to it is usually this kind of PID that the speaker has in mind. (2) Post abortal PID (septic abortion, 6.6a). (3) Infected obstructed labour (18.4). (4) Puerperal sepsis (puerperal PID', 6.7). (5) Sepsis after Caesarean section (postcaesarean PID', 6.8).
A patient with postmenstrual PID' is not pregnant, she has suffered no birth trauma and there are no infected products of conception. She may however have an IUD in her uterus, which increases the risk of serious infection and may delay recovery. This kind of PID is seldom fatal, and never causes septic thrombophlebitis. All the others are dangerous, and commonly kill her. Postabortal peritonitis is particularly deadly and has a mortality of 50%. A girl of 17 may waste away with a bowel fistula like ]]a terminal cancer patient, as did the patient, Grace', described below.
Fig. 6-9 PELVIC SEPSIS. A, infection spreading from the uterus to cause peritonitis. Infection can also spread as an infected thrombus (thrombophlebitis). B, infection of the connective tissue beside the uterus (parametritis). Infection may spread into the broad ligament, round the vagina or uterus, or up into the loin. C, a collection of pelvic pus. D, salpingitis. After Garry MM et al. Obstetrics Illustrated', pp. 319[nd]320. E and S Livingstone. with kind permission. The organisms responsible for PID may be: (1) Sexually transmitted[md]gonococci (either penicillin-sensitive, or PPNG', penicillinase-producing Neisseria gonorrhoeae), mycoplasma, or chlamydia (both less common). (2) The normal flora of her gut and vagina[md]coliforms, various anaerobes, and rarely actinomyces. The latter organisms (and sometimes even the former) live harmlessly in the vagina and cervix, and only cause disease when the barriers to spread are removed by: (a) Abortion or delivery (very common). (b) Menstruation. Or, (c) some medical intervention, such as a D and C' (not uncommon), the insertion of an IUD (common but usually mild), or a hysterosalpingogram (rare).
Many gonococci, and typically all chlamydia and mycoplasma are sensitive to tetracycline. But when PID follows pregnancy or an abortion, it is caused by a mixture of organisms, including anaerobes, for which she needs metronidazole with chloramphenicol. By the time you see her, secondary invaders are likely to be present, whatever the primary cause of her infection.
HIV makes a mild genital infection more likely to progress to PID, so if yours is a high AIDS area, test for her HIV if you can.
The clinical manifestations of pelvic sepsis are wide. They range from an otherwise symptomless infertility caused by blocked tubes, to generalized peritonitis, septicaemia and septic shock, with everything between these two extremes. Like a fire, PID can be of any degree of severity, from smouldering to fulminating. Also, like a fire, it can die down, only to light up again later. So you will see: (1) Early acute cases who may not become infertile, if you treat them early and energetically. (2) An occasional fulminating case with peritonitis and shock. This can be an early acute case which is particularly severe, or it can be due to a tubo-ovarian abscess, which has previously caused only minor symptoms, bursting into the peritoneal cavity. (3) Chronic cases. (4) Chronic cases with a flare-up.
The typical acute case of postmenstrual PID has fever, bilateral lower abdominal pain, and tenderness, but seldom any rigidity. She usually also suffers from frequency of urine, dyspareunia, heavy or prolonged periods, and usually also has a vaginal discharge (see below). She may not admit to all these symptoms, especially if she is a young unmarried girl. On pelvic examination, she is usually equally and acutely tender in both her vaginal fornices (unlike an ectopic pregnancy, in which she is usually only tender in one of them). Her pain may be so intense that you have to repeat the examination after you have given her an analgesic. She may also have a lower abdominal mass, vomiting, fever and a raised ESR.
Her symptoms are usually mild, but can be severe with signs of peritonitis and occasionally septic shock. Acute cases are often atypical, either because the disease is mild, or because it has been modified by previous treatment.
The typical chronic case complains of infertility, and pelvic pain, often with dyspareunia, poor general health, and much misery. The diagnosis may be difficult, and the differential diagnoses include psychosomatic pain.
You can usually treat a patient non-operatively. Occasionally, you will need to drain pus. Unfortunately, once PID has become chronic, she may have recurrent pain, and if she is educated her threshold to it is likely to be lower. Don't operate on chronic PID unless you have to, because once it has been present for more than a few weeks, her pelvic organs will be so densely stuck to one another that freeing them will be difficult and dangerous[md]you can easily injure her gut. If you have to operate, do so on the indications given below, and be conservative. Leave her pelvic organs intact unless she has a tubo-ovarian abscess. Removing this can be difficult, so open it and drain it. If necessary, and you are sufficiently skilled, return later to remove her tubes in the chronic phase, or (better) persuade someone else to. If she can be left with her uterus and some ovarian tissue, she will continue to menstruate.
GRACE NYRIENDA (17) was admitted with vaginal bleeding and fever, having attempted to procure an abortion on herself at 16 weeks. Her cervix was wide open, the products of conception were visible, and there was a foul discharge. She was treated with antibiotics and her uterus was evacuated. A few days later she was very ill with a distended abdomen. Three litres of thin pus were washed out of her peritoneal cavity and tetracycline was instilled. There was no perforation in her uterus. She was treated with more antibiotics, intravenous fluids, and nasogastric suction. Two weeks later she was still febrile and very ill. A second laparatomy was done to drain residual abscesses. Chronic sores developed at the sites of the drainage tubes, which continued to discharge pus. She did not eat well, and vomited from episodes of subacute obstruction, but was not well enough for a third laparotomy. Three months after admission she died extremely wasted. LESSONS (1) This is a typical history; there were no obvious mistakes in her treatment. Often, there is nothing you can do. (2) Any abortion, particularly a procured one, is dangerous at 16 weeks. Fig. 6-10 A PLAN FOR MANAGING PID. Stage One: the patient has no peritoneal irritation, and no pelvic mass. Stage Two: she has peritonitis as shown by bilateral lower quadrant rebound tenderness. Stage Three: she has a mass in her adnexa (tubo- ovarian abscess) or pouch of Douglas. Stage Four: she is very ill indeed, as after the rupture of a tubo-ovarian abscess. This regime differs slightly from that in the text. After De Mulder X, Pelvic inflammatory disease in Zimbabwe', Tropical Doctor 1988;2:85.
PID [s7](postmenstrual PID') This is the patient with PID who has not recently aborted (6.6a), or delivered (6.7), or had a Caesarean section (6.8), and who typically has postmenstrual PID', commonly either gonococcal, chlamydial, or mycoplasmic. For the drainage of a pelvic abscess, see Section 6.5. For the management of adhesions see also Section 10.7.
[+20]ACUTE PID DIFFERENTIAL DIAGNOSIS. Acute postmenstrual PID' has mostly to be distinguished from other causes of acute lower abdominal pain, including appendicitis (12.1) and a urinary infection (10.2). The main gynaecological differential diagnosis is a ruptured ectopic pregnancy (16.6). Fixity of her pelvic organs on vaginal examination is no help in distinguishing between PID (common), tuberculosis (uncommon), and endometriosis (rare), because they all do this.
Suggesting a ruptured ectopic pregnancy[md]a Yes' answer to the following questions: (1) Is she more than slightly anaemic? (2) Has she missed one or more periods by more than a few days? This is often followed by a small loss of dark or brownish blood vaginally. (3) When you examine her vaginally, is she more tender on one side than on the other? (4) Has she a mass on one side? In a subacute ectopic pregnancy the Fallopian tube mass is unilateral, but the pelvic haematocele of a chronic ectopic pregnancy usually feels as if it is in the midline. (5) Is she afebrile? An ectopic pregnancy does not usually cause fever, whereas acute or subacute PID usually does.
Suggesting a twisted ovarian cyst: a mass, no fever, and colicky lower abdominal pain, sometimes with vomiting.
If you are not sure of the diagnosis, and have a laparoscope (15.4), look for red, sticky, and oedematous tubes.
CAUTION ! (1) Her vaginal discharge is not proportional to the severity of her PID. Candida and Trichomonas cause a profuse discharge, but do not usually cause PID. Gonococci and Chlamydia cause a less obvious mucopurulent discharge. (2) Expect your diagnosis to be wrong in about 20% of patients.
MANAGEMENT. You can usually treat her as an outpatient. Admit her if: (1) She is too ill to go home, especially if she has: (a) bilateral lower quadrant rebound tenderness (indicating peritonitis), (b) a mass or (c) shock. (2) You cannot exclude an acute surgical condition, especially an ectopic pregnancy. (3) Outpatient treatment has failed. (5) She is unlikely to take her drugs or return for follow up.
ANTIBIOTICS. Cervical smears and cultures are of little help in choosing an antibiotic, because the organisms in her cervix may not be those which are causing the infection elsewhere. The absence of gonococci in a cervical smear does not exclude gonococcal infection. Usually, you will need to treat her blindly with a broad-spectrum antibiotic. If possible follow up and treat her partners.
Give her tetracycline 500 mg and metronidazole 400 mg, both 4 times daily for 7 to 10 days (2.9). Doxycycline 100 mg twice daily is better than tetracycline but is more expensive. Encourage her to complete the course. She is likely to stop if she feels better. Also, give her an analgesic.
If she is very ill with signs of spread outside her uterus, she needs parenteral antibiotics (2.9) in high dose against a wide range of organisms. Give her: (1) Benzyl penicillin 1.2 g 6-hourly. And, (2) give her chloramphenicol 1 g intravenously immediately followed by 500 mg intravenously 6- hourly. And, (3) give her gentamicin 80 mg intravenously 8- hourly. Or, give her intravenous chloramphenicol and metronidazole.
IF SHE FAILS TO RESPOND TO NON-OPERATIVE TREATMENT, as shown by feeling better, a falling temperature, less pain, and a reduction in the size of the mass; and particularly if she gets worse: (1) Have you made the right diagnosis? (2) Is there a collection of pus somewhere which needs draining? (3) Is she on the right antibiotic in the right dose? If you do decide that this is the problem, make sure you have excluded a wrong diagnosis or collections of pus.
Many patients with acute PID have a mass of matted viscera (as distinct from an abscess which needs drainage). This may take 6 weeks to resolve, but there is no point in continuing antibiotics for more than 2 weeks. If she is not well and has a spiking temperature after 6 weeks, she probably has an abscess which needs draining vaginally or suprapubically.
CAUTION ! Don't change antibiotics unless: (1) You have given them for at least 3 days. (2) You have carefully reviewed her. (3) You are sure that your original diagnosis of PID is correct. (4) You are reasonably sure she does not have a collection of pus anywhere.
LAPAROTOMY [s7]FOR ACUTE PID INDICATIONS. Refer her if you can, this is not an easy operation. She has a significant chance of dying. There are three indications for a laparotomy:
(1) The diagnosis is in doubt, and there is a possibility that she might have an ectopic pregnancy or appendicitis, for example.
(2) After 48 hours of antibiotic treatment for PID (particularly after a septic abortion, 6.6a), she is not improving. Instead, her pulse continues to rise, her temperature is maintained, and there are signs that peritonitis is spreading.
(3) She has sudden generalized peritonitis with shock due to rupture of a tubo-ovarian abscess. This may be spontaneous or it may follow a vaginal or rectal examination. If her history is suggestive, resuscitate her and operate immediately. She is in grave danger.
RESUSCITATION. Give her intravenous fluids (A 15.3). She may need 3 or 4 litres of glucose saline during the first 24 hours. She may bleed considerably from the raw surfaces that will form when you free the adhesions between the loops of her gut, so have two units of blood cross-matched. If she is seriously ill, she is in danger of renal failure, so insert an indwelling catheter and monitor her urine output. Pass a nasogastric tube.
PERIOPERATIVE ANTIBIOTICS. If she is not already on tetracycline and metronidazole, start these before the operation. Or, give her chloramphenicol and metronidazole (2.9).
ANAESTHESIA. (1) General anaesthesia with intubation (A 10.1). (2) If she is very sick, you can if necessary operate under local infiltration anaesthesia (A 6.9) and heavy premedication with pethidine and diazepam (A 5.2), but this will be unpleasant for you and for her. Also, local anaesthetics do not work well in the presence of infection. (3) Ketamine (A 8.1) with local infiltration anaesthesia of her abdominal wall.
INCISION. Make a lower midline incision (9.2) and extend it above her umbilicus if necessary. Here are some of the things you may find. Also be prepared on occasion to find some quite unexpected condition such as a perforated typhoid ulcer (31.8).
If the infection is limited to her pelvis, examine her upper abdominal cavity before you explore her pelvis and disturb the adhesions, which are limiting the spread of infection. Examine her subphrenic and subhepatic spaces, and her paracolic gutters; look for abscesses between the loops of her small gut as far as you can reach them. If you find pus, deal with it as in Sections 6.2, 6.3 and 6.4. If you find dense adhesions, see also Section 10.7.
If you don't find pus in her upper abdomen, carefully protect the upper uninfected part of her abdominal cavity with large abdominal packs. Slowly and methodically divide the adhesions between her gut and her uterus, and look for pus. Divide the adhesions round her tubes and ovaries, and release the pus you find there. Try to get right down into her pouch of Douglas. There is usually no need to remove her tubes or ovaries, however diseased they may look. The tubes have a double blood supply which prevents them becoming gangrenous, and they are not connected to a contaminated viscus like the appendix.
When you find her fundus push your fingers down behind it, between her tubes, which will almost meet in the middle. You need not fear perforating her gut here. Gradually work your fingers down below her tubes. Free them from her gut from below upwards.
Remember her anatomy: it is always the same. Both tubes will be stuck down behind her uterus, over the top of each ovary. Her rectum and colon will be adherent from below upwards to the back of her uterus, and then to both her tubes. Loops of small gut and omentum will have stuck to them on top. If you can find her fundus you will know where you are.
Don't panic when you find a mass of adherent gut and omentum. It will always come clear in the end. First get down to her fundus by lifting off her gut and omentum. Divide all adhesions and release all pockets of pus.
CAUTION ! Don't tear her gut. Avoid doing so by going slowly, and squeezing and pinching the plane of cleavage between your fingers (10.7). Cut dense adhesions with scissors.
If she has generalized peritonitis, suck away as much pus as you can, then suck out her paracolic gutters. Make sure you release any collections of pus under her abdominal wall, between her large gut and her abdominal wall, and under her diaphragm and her liver (subphrenic and subhepatic spaces). Bring out her whole small gut over its full length in stages. Break down adhesions between loops of gut, by careful blunt dissection, to release the many collections of pus between them. Then go to her pelvis, and proceed as above for a localized pelvic infection.
If you find she has a septic abortion, you will have to make the difficult decision as to whether or not to do a hysterectomy. Assess the state of her uterus and adnexa. By the time she has generalized peritonitis, hysterectomy is probably best: (1) The main indication for it is a perforated septic abortion. (2) How many children has she? If she is young and has no children, losing her uterus will be a major disaster. Even if you leave it, she will probably be infertile. (3) How skilled are you? If you are skilled this favours hysterectomy. A subtotal operation will be enough, but it will be dangerous. Occasionally, you may be able to avoid hysterectomy and do a salpingo-oophorectomy if generalized peritonitis seems to originate in an abscess in one of the adnexa (uncommon). Usually, all you need do when this happens is to drain pus and leave a tube in the abscess.
If you find that there is acute inflammation in her pelvis, and perhaps elsewhere without much pus, the infection is very early and she is lucky. Wash out what pus there is. Close her abdomen and continue chemotherapy.
If she has a ruptured tubo-ovarian abscess, leave the tube in and insert a drain.
LAVAGE AND DRAINS depend on the extent of the sepsis you found:
If the pus was localized to her pelvis, wash it out of her pelvis only (6.2), before you remove the packs protecting the rest of her peritoneal cavity. Place two tube drains in her pouch of Douglas, and bring them out through stab incisions, lateral to her rectus muscles.
If she had generalized peritonitis, drains do not work well, so wash out her whole peritoneal cavity as in Section 6.2.
CLOSURE. Close her abdomen as a single layer and leave her skin open for secondary closure (9.8). This is better than inserting tension sutures.
DIFFICULTIES [s7]WITH ACUTE PID Be prepared for small gut fistulae (9.14), and a burst abdomen (9.13), especially if abdominal distension persists for some time postoperatively.
If she has a mass and you are not sure if she has a RUPTURED ECTOPIC OR A PELVIC ABSCESS, do a culdocentesis (16- 6) under general anaesthesia. If you find pus, drain it through her vagina. If you find blood which fails to clot, do a laparotomy (16.7).
If she HAS AN IUD IN AND PRESENTS WITH WITH PAIN, you will not find it easy to decide if her IUD is causing her pain, or if PID is causing it. If her symptoms are not too severe, and her cervix is merely a little tender, see if she will settle quickly with antibiotics and an analgesic. If she settles, leave her IUD. If you always remove an IUD because it causes a little pain and tenderness, you will remove too many. If she is febrile and very tender, give her antibiotics for 24 hours and then remove her IUD.
CAUTION ! Don't remove her IUD immediately. Removing it from her acutely infected cervix will be very painful.
If you find ACUTELY INFLAMED TUBES (SALPINGITIS), when you expect to find something else, leave them, her infection will settle if you give her an appropriate antibiotic. Do a peritoneal toilet and close her abdomen. Unlike an appendix, which you must remove (12.1), her tubes will not become gangrenous, or form a faecal fistula, or leak faeces into her peritoneal cavity.
If you enter her abdomen, expecting to find pus, but FIND LITTLE PUS OR NONE, and few signs of inflammation, examine her pelvic organs and particularly her infundibulopelvic ligaments (20-17). One or both may be thickened and oedematous, and the thickening may extend under her ovaries to her uterus. If so she has SEPTIC THROMBOPHLEBITIS of her ovarian veins (not uncommon). If you find nothing, the thrombophlebitis is probably in a uterine vein which is not so easily seen. If you don't find anything else, particularly any pus, close her abdomen. Continue with antibiotics in high doses. If possible, 24 hours after the operation start her on intravenous ]]heparin 5000 to 10,000 units by bolus intravenous injection 6-hourly, controlled by estimating her clotting time and lengthening it to about 15 minutes. Continue this for at least 2 weeks. Watch carefully for abnormal bleeding, particularly from the abdominal incision, or her urinary or intestinal tracts. She should improve quite quickly.
If you find DISSEMINATED YELLOWISH-WHITE NODULES throughout her pelvic cavity, or a localized infection in her pelvis with nodules on her tubes and perhaps a CASEOUS ABSCESS, suspect that she has TUBERCULOSIS. Take a biopsy and send this for histology.
If you ACCIDENTALLY TEAR HER PELVIC COLON, what you should do depends on the size of the tear and where it is. If it is small, oversew it. If it is large, either close it and make a defunctioning colostomy (9.5) higher up, or divide her colon, close the distal end, and bring the proximal end out as a terminal colostomy (Hartmann's procedure, 9.5).
If there is PERSISTENT SEPSIS in her peritoneal cavity, in spite of repeated attempts at drainage, she is likely to go steadily down hill and die, after several months of great suffering (see the story of Grace above).
[+20]CHRONIC PID DIAGNOSIS. Feel for tenderness of her uterine adnexa on bimanual examination, and for tender masses.
THE DIFFERENTIAL DIAGNOSIS includes urinary tract infection, endometriosis (rare in the developing world), and pelvic tuberculosis (uncommon).
ANTIBIOTICS. Give these as for acute PID for 10 to 14 days. If she has a recurrent infection, consider giving her three courses starting on the first days of successive menstrual periods.
If she improves, she feels better, her pain is less, and her mass disappears over 1 to 3 months.
If she does not improve, either your diagnosis is wrong, or she has a collection of pus, perhaps a chronic tubo- ovarian abscess or a pyosalpinx. Treatment is difficult. Refer her.
DIFFICULTIES [s7]WITH CHRONIC PID If, on laparoscopy or laparotomy, you see BLUISH OR BROWN NODULES on the surface of her peritoneum and particularly on her uterosacral ligaments, surrounded by puckering, suspect ENDOMETRIOSIS. You are most likely to see such nodules on her uterosacral ligaments, in her pouch of Douglas, on her ovaries, on the posterior surface of her broad ligament, or on the fimbrial ends of her tubes. Refer her. Or, if she has pain give her a non-cyclical progestogen to suppress menstruation, such as norethisterone 10 mg daily starting on the 5th day of the cycle (increased if spotting occurs to 25 mg daily in divided doses to prevent break-through bleeding) for at least 6 months. Or, give her Depo-Provera 50 mg weekly or 100 mg every 2 weeks for 3 months.
If she is a YOUNG WOMAN WHO COMPLAINS OF INFERTILITY, menstrual irregularity, and chronic pelvic discomfort, TUBERCULOSIS (29.5) is a possibility.
If she has an IUD in and presents with UNILATERAL SIGNS and perhaps a HARD TUBO-OVARIAN MASS, suspect ACTINOMYCOSIS (rare) as the result of the introduction of the IUD. Confirm the diagnosis by biopsy at laparotomy. She will recover on treatment, and may possibly become fertile. Give her 2 megaunits of benzyl penicillin 6-hourly for 2 weeks, and then oral penicillin V 250 mg 6-hourly for 8 weeks.
If she has chronic PID and is worried about INFERTILITY, you can assure her that removing masses will not make her fertility better or worse, because she is probably incurably infertile already.
If she has chronic PID and is worried about PAIN but is not worried about having any more children, unilateral or bilateral salpingectomy without hysterectomy is usually possible. This is difficult, so don't attempt it yourself unless referral is impossible and you have considerable operative experience.
Fig. 6-11 PID AND PELVIC TUBERCULOSIS. A, acute salpingitis with swollen congested tubes and pus leaking from the ostium. B, chronic salpingo-oophoritis with the tubes and ovaries densely bound by adhesions. C, a tubo-ovarian cyst. D, a tuberculous pyosalpinx. E, a hydrosalpinx. From Young, James, A Textbook of Gynaecology' (5th edn, 1939). A and C Black.
Septic abortions
If a patient with an incomplete abortion has fever and pus discharging from her cervix, the products of conception have become infected. This can follow a neglected spontaneous abortion, or it can follow unskilled interference. Fortunately, the uterus is usually a good barrier to the spread of infection, but it does sometimes spread as pelvic cellulitis or peritonitis. You can usually treat her without a laparotomy, although you will usually need to evacuate her uterus. If she has peritonitis you will have to open her abdomen.
The diagnosis is usually easy[md]if her history is clear and she is obviously pregnant. Unfortunately, she may be so frightened that she will deny having tried to procure an abortion, even when she is very ill. The only way to avoid a misdiagnosis is to remember that any acute pelvic inflammation in a woman of childbearing age may be the result of an abortion.
Try to control both haemorrhage and infection before you empty her uterus, usually after about 24 hours on antibiotics. Rarely, a hysterectomy may be the only way to save her life. The great dangers are septic shock (53.4) and renal failure (53.3).
IF SHE IS OF CHILDBEARING AGE, IS HER PELVIC INFLAMMATION THE RESULT OF AN ABORTION?
SEPTIC ABORTION See also Section 6.6.
THE DIAGNOSIS should not be difficult. A patient becomes ill and febrile ([mt]38[de]C) after an abortion. She has a foul vaginal discharge, and sometimes frank pus. Examine her vaginally in the ward. If you have the necessary facilities, start by taking an endocervical swab for culture aerobically and anaerobically (if possible). This is better than a high vaginal one. Take it yourself: if you leave it to a junior nurse, it is likely to be a low' one. Then use your fingers to remove any of the products of conception, which will come away easily.
Examine her bimanually. Her uterus is tender, there is tenderness on either side, perhaps with a mass. Sometimes she has local or general peritonitis.
Look also for anaemia, jaundice (caused by septicaemia) and chest signs (septic emboli from thrombophlebitis). Measure her haemoglobin, and take blood for grouping and cross-matching. If possible take blood cultures.
Your main concern will be to know how far infection has spread, and if you should open her abdomen.
If her pulse is over 120 a minute, the infection has probably spread beyond her uterus.
If moving her cervix causes her great pain and her lateral fornices are hot, thickened, and tender, perhaps with a mass, the infection has spread to her pelvic connective tissue (parametritis, uncommon).
If you are uncertain about the diagnosis and she is very sick, resuscitate her, start her on antibiotics, take her to the theatre and aspirate her posterior fornix. A seriously infected uterus can be silent, apart from a very sick patient.
If her history suggests that her uterus has been perforated with some instrument, her prognosis is worse. If it is leaking pus into her peritoneal cavity, you may ultimately have to do a hysterectomy.
RESUSCITATE HER as in Section 6.6. If she has lost blood, transfuse her as necessary (53.2, 53.4). If you have no blood, give her Ringer's lactate or 0.9% saline, or a plasma substitute.
ANTIBIOTICS. Try to prevent the infection spreading beyond her uterus. This risk is greatest when you evacuate it. So always cover evacuation with perioperative antibiotics. They will not control the infection, if infected products of conception remain inside her uterus. So empty it, and don't expect to cure her until you have done so.
If she is not very ill, and there are no signs that infection has spread beyond her uterus, a single broad spectrum antibiotic, such as ampicillin may be enough: (1) Give her ampicillin 500 mg intravenously 4 to 6-hourly. Or, (2) give her benzyl penicillin 600 mg intravenously 6-hourly, with streptomycin 1 g daily intramuscularly. Or, instead of the streptomycin give her gentamicin 80 mg intravenously 8-hourly.
If she is very ill, with signs of spread outside her uterus, she needs parenteral antibiotics[md]see Section 6.6.
ANALGESICS. Give her pethidine.
CONTROLLING BLEEDING. Give her ergometrine with oxytocin (Syntometrine' 1 ml) intravenously, or ergometrine alone.
EVACUATIION. Opinions differ as to when this should be done. We advise that you do it a few hours after starting antibiotics, and never later than 24 hours after. If she is bleeding seriously, do it immediately. Follow the method in Section 16.2. Her uterus will be infected and soft, so be especially careful not to perforate it. Use a blunt curette. Continue antibiotics after evacuation.
POST-EVACUATION MANAGEMENT. Watch her carefully, especially her urine output. Several things may happen during the next few days. If she is seriously ill, an important decision will be whether or not she needs a laparotomy.
If all is well, she should improve dramatically, and her fever should go in 48 to 72 hours.
If she has not started to improve 24 hours after evacuation, but signs of peritonitis are not obvious, she probably has a pelvic abscess. Take her to the theatre and aspirate her posterior fornix (6.5). Avoid her lateral fornices, or you may injure her ureters or her uterine arteries. If you aspirate pus or blood-stained fluid, drain it through her posterior fornix, as in Section 6.5.
If she has not started to improve 24 hours after evacuation, and signs of generalized peritonitis are obvious (pain, tenderness, rigidity, and abdominal distension), she is in serious trouble. Her uterus may have been perforated by an abortionist, or he may have injected some harmful fluid into her peritoneal cavity. She needs a difficult laparotomy. Refer her if you can. Improve her general condition as best you can. Rehydrate her, if necessary, restore her haemoglobin to at least 80 g/l by transfusion, and have at least 2 units of blood available. Then do a laparotomy as in Section 6.6.
DIFFICULTIES [s7]WITH SEPTIC ABORTIONS See also Sections 6.6 and 16.2 (inevitable abortions).
If you are NOT SURE IF SHE HAS PERITONITIS or not, consider waiting 24 hours. If necessary, aspirate her posterior fornix with a needle in the theatre.
If you PERFORATE HER UTERUS when you evacuate a septic abortion, there is no easy answer. If you stop and send her back to the ward incompletely evacuated, she is in danger. If you complete the evacuation, you may spread the infection further. This also is dangerous. As a general rule, if you perforate a pregnant uterus, complete the evacuation as best you can, then do a laparotomy and repair her uterus with a single layer of interrupted silk sutures. For the accidental perforation of a non-pregnant uterus, see Section 20.3.
If her URINE OUTPUT FALLS below 30 ml/hour, her kidneys are probably failing, so see Section 53.3.
If she has signs of SEPTIC SHOCK, treat it (53.4). The signs to watch for are an alert patient with a blood pressure below 80 mm, and a subnormal temperature.
If you feel CREPITATIONS (bubbles of gas in the tissues), suspect GAS GANGRENE and look for gas shadows radiologically. Treat her as in Section 54.13.
Puerperal sepsis
After childbirth a patient's genital tract has a large bare surface, which can become infected. Infection may be limited to the cavity and wall of her uterus, or it may spread beyond to cause peritonitis (6.2), septicaemia, and death, especially when her resistance has been lowered by a long labour or severe bleeding. If she is more fortunate, her infection may be walled off by her gut and omentum. She may have a pelvic abscess with pus in her pouch of Douglas, or she may have pus high in her pelvis or in her lower abdomen.
If sepsis is localized, only her lower abdomen is distended, she has guarding in both her iliac fossae, and an ill-defined tender mass arising from her pelvis. She may have hyperactive bowel sounds. Vaginally, she shows signs of recent childbirth or abortion, and may have infected lacerations. Her cervix is open and tender, painful on movement, and may be drawn up behind her symphysis. Her pouch of Douglas may be thickened or swollen, but you cannot feel a fluctuant mass vaginally. Her uterus and appendages form a mass which is difficult to define because of their tenderness.
If sepsis is generalized, she is weak, with anorexia, fever (perhaps with rigors), a rapid thready pulse, a low blood pressure and generalized abdominal pain. Her abdomen is uniformly distended, tympanitic, silent, and acutely tender. She may have a visible mass extending up to her umbilicus; you may have to pass a catheter to make sure that it is not merely a distended bladder. She cannot walk. She may have diarrhoea until peritonitis causes ileus and this causes constipation and vomiting.
PUERPERAL SEPSIS See also Sections 6.6 and 6.6a.
RESUSCITATE HER, if necessary, as in Section 6.6.
GIVE HER ANTIBIOTICS, as in Section 2.9. (1) Give her chloramphenicol and metronidazole. Or, (2) give her: ampicillin 500 mg 6-hourly for 7 days, and metronidazole (2.9). If she is very ill, she must have metronidazole either intravenously, or as suppositories, or tablets rectally (2.7). Too little chloramphenicol will be excreted in her breast milk to harm her baby. Or, give her gentamicin, or kanamycin.
MONITOR HER daily for signs of the spread of infection.
MANAGE HER like this:
If she continues to bleed, she may have retained pieces of placenta. This is a common cause of puerperal sepsis, which will not resolve until her uterus is empty. Give her antibiotics and curette her 24 hours later with great care! Use the largest curette which will be less likely to perforate her uterus. Curetting a large, soft, infected uterus is dangerous.
If her uterus is enlarged and tender, with a closed cervix as the result of scarring or carcinoma, it may be full of pus (pyometra, 32-21). This can occur 2 weeks or more after delivery. Drain her cervix with Hegar's dilators, 10 Ch is usually enough.
If she has a definite swelling at one side of her uterus, she has parametritis.
If she has generalized peritonitis without any localizing signs, make a muscle splitting incision as for appendicectomy in an iliac fossa. Open her peritoneum, sweep gently with your finger, and insert a sump sucker. Up to a litre of thin pus will probably escape. If you enter an abscess cavity, gently free any adhesions and open up all loculi. Wash out her peritoneum (6.2), and then instil tetracycline 1 g in a litre of saline.
If her fever recurs after initial improvement, there is more pus somewhere which must be drained, either through the same incision or another one. If you fail to drain a subphrenic abscess (6.4), she will die.
If she recovers from the acute episode, but is left with a mass, she may eventually need a need a full laparotomy, with the separation of adhesions and the removal of a tubo- ovarian mass. Refer her if you can.
Fig. 6-12 PUERPERAL SEPSIS. There is septic thrombophlebitis. Septic emboli are spreading through the ovarian and internal iliac veins to cause septicaemia and and abscesses in the lungs and kidneys. Adapted from an unknown source..
Infection following Caesarean section
Peritonitis (6.2) may follow any obstructed labour, or an infected Caesarean section, and is common after rupture of the uterus. If a patient dies she will probably do so because you did not anticipate infection, or because you opened her abdomen much too late. She is likely to be infected: (1) If her labour is abnormally long, and the longer it lasts, the greater the risk. (2) If her baby is dead. (3) If her membranes rupture early and her liquor becomes infected. (4) If your sterile procedures are poor. In any of these conditions, anticipate infection and try to prevent it.
INFECTION [s8]AFTER CAESAREAN SECTION See also Section 6.6, 6.6a and 6.7. For vaginal bleeding due to infection (secondary PPH) see Sections 18.10 and 19.11b.
TRY TO ANTICIPATE INFECTION. If you expect that infection will follow Caesarean section: (1) Give the patient the perioperative antibiotic regime in Section 2.9, and (2) consider doing an extraperitoneal Caesarean section (18.13).
If you do not do an extraperitoneal operation, put a pack in each paracolic gutter, and one or two above the incision between her uterus and her abdominal wall. Make a small incision in her uterus first, and suck out her infected liquor and meconium. After delivering the placenta, mop out her uterus, and remove all remnants of membranes and some decidua. After closing her myometrium, and before closing her uterine peritoneum, wash out her pelvis 2 or 3 times with warm saline or water. Repair her peritoneum, control bleeding meticulously, remove the packs and clean up very, very carefully.
Alternatively, and probably less satisfactorily, after doing an ordinary lower segment operation, insert corrugated rubber drains, about 3 fingers width, in each of her paracolic gutters, and lead them out loosely through incisions in her abdominal wall. Cut her peritoneum round these drains, and don't merely stretch it. Insert another good wide drain suprapubically. Make sure there is no remaining amnion in her uterus to prevent free drainage through her cervix. Some surgeons insert insert a fourth drain through her cervix into her vagina.
After you have done the Caesarean section close her peritoneum and rectus sheath with a single layer of interrupted through-and-through sutures of stainless steel wire, or strong monofilament, 1 cm apart, taking 1.5 cm bites each side of the wound (9-21). Don't close her skin immediately; instead, close it by delayed suture (9.8).
CAUTION ! Watch carefully for the first signs of infection[md]fever, and a large soft tender uterus with tenderness deep in her flanks.
IF INFECTION OCCURS, it may take the following forms. It can also cause secondary haemorrhage (3.10), or sterility (15.2). If you left packs or swabs in her abdomen, low-grade peritonitis may follow and obstruct her gut. See also 6.6D.
If pus forms around the wound (9.12), it may discharge through the scar into the cavity of her uterus. Infection may resolve, or you may need to drain pus suprapubically.
If she develops a pelvic abscess, manage it as in Section 6.3.
If pus forms in her peritoneal cavity and spreads upwards, manage her as for peritonitis (6.2, 6.6). Continue intravenous fluids, and gastric suction. Open her abdominal wound, and suck pus from all the cracks and crevices. If her general condition is poor, do this under local anaesthesia. You will probably find that her uterus is totally disrupted, so it is hopeless to try to repair it, and almost certainly fatal to try to remove it. If you do decide to do a hysterectomy, a subtotal operation will usually be enough[md]commonly with the removal of both adnexa, but retain one if you can.
Wash out her abdomen and instil tetracycline (6.2). Consider inserting four drains, one in each paracolic gutter, one down to the wound, and one through her cervix into her vagina, making sure that this last is not occluded by amnion. Close her abdomen with interrupted wire sutures and leave her skin wound open.
If she has a tender suprapubic mass, make a 5 cm muscle-splitting incision over it, as for appendicectomy. Open her peritoneum, insert a sump sucker, and sweep your finger round the inside of her peritoneum as far as you can reach. A litre of thin pus may escape. Stitch in a large drainage tube.
If she has signs of pus somewhere' (a hectic fever, malaise, and anorexia), but there are no obvious signs of it, suspect that she has a subphrenic abscess. This is a common late complication, and is likely to kill her if you don't drain it (6.4); so may multiple abscesses between loops of her gut (6.3).
If her abdominal wall bursts, and exposes her uterus, repair it.
If her fever recurs, and there are signs that more pus is collecting, do another drainage operation.
