REFERENCES

1. ROf'abeck CH. Tourniquet-induced nerve ischemia: An experimental investigation.J T/8umi! 1980; 20(4): 280-286.

2. Bruner JM. Safety fal;tors in the use of the pneumatic tourniquet for hemostaSlsin SW'gllf)'of the hand. J Bone Joim Surg (Am) 1951; 33: 221-224.

3. Eckhoff NL Toumiquet paralysis_ A plea for the extended use of the pneumatictoutniquel.l..ancet 1931; Aug.: 343-34.5.

4. PaJetta FX, Winman V. Ship AG. Prolonged tourniquet ischemia of eXll'emities. Anellperimental study on dogs. J Bone Joinr Surg (AmJ 1960; 42: 9<:5-950.

5. Sn.aw WUgis EF. Observations on the effects of toutniquet iSChemia. J Bone JointSurg (Am) 1971; sa: 1343-1346.

6. Gersoff WK. Ruwe P. Jokl P. Panjabi M. The effect of lourniquet pressure onmuscle function. Am J SportS Med 1989: 17(1): 123-127.

7. Nitt AJ, MatuJioni$ OH. Ultrastructural changes in rat peripheral nerve followingpneumatic tourniquet comPl'"ession. J Neurosurg 1982; 57; 660-666.

8. Dobner JJ, NilZ AJ. Postmeniscectomy tourniquet palsy and functional $equelae.Am J Sports Med 1982; 10: 211-214

9. Nitz AJ, Oobner JJ. Upper extremity tourniquet effects in carpal tunnel r&lease. JHand Surg (Am) 1989; 14: 499-504.

10. Gutin B, Warren A. Wickiewicz T, er al. Does tourniquet use during anteriorctlJc;ate ligament surgery interfere with postsurgical recovery of function? Areview of the Uterature. ArtJrroscopy 1991; 7(1): 52-56.

11. Saunders KC, Loui$ OL, Wainganlen SI. Waylonis GW. Effect of tourniquet timeon postoperative quadriceps function. Clin Orthop 1979; 143: 194-199.

12. Miller SH, Price G. Buck O. et al. Effects of tourniquet ischaemia anopostischaemic !dema on muscle metabolism. J Hand 5urg 1979; 4(6): 547-555.

13. KJenerman L, Biswas M, Hulands GH, Rhodes AM. Systemic and local effects ofthe application of a tourniquet. J Bone Joint Surf; (Br) 1980; 62: 385-388.

14. Gardner VC, CaiOZlO UJ, Long ST, er al. Contractile properties of slow and fastmuscle following tourniquet iSChemia. Am J Sports M~ 1984; 12(6): 417-423.

15. Heppenstall RB, SCOll R, Sapega A. er al. A comparative study of the toleranceof skeletal muscle to ischemia. J Bone Joint SUr9 (Am) 1986: 68(6): 620·828.

t6. Gregory MA. Mars M. Serial morphological changes in primate skeletal myofibresafter 3 hours of ischaemia and 24 hours of reperlusion. 5 Air Med J 1992: 81:473-478.

17. HeppenstaJl RB. Ba/derston R. GOOdwin C. Pathophysiological ellects disra! to atourniquet in the dog. J Trauma t979; 19(4): 234-238.

18. Irvlng G. The pneumatic tourniquet: A literature review and studies to improve Itssafety in clinical practice. M.Sc.Med. thesis, Unive~ity Of Cape Town. 1981.

19. Reid HS, Camp RA. Jacob WHo Tourniquet hemostasis - a clinical study. ClinOrthop 1983; 177: 231-234.

20. Shaw JA. Murray 00. The relationship between tourniquet pressure andunderlying soft·tissue pressure in the thigh. J Bone Joint 5urg (Am) 1982; 64(8):1148-1152.

21. Calderwood JW, Dickie WR. Tourniquet pares's complicatin9 tendon grafting.Hand 1972; 4: 53-55.

22. Fry O. Inaccurate tourniquet gauges (Letter). BMJ 1972: 1: 511.

Accepted 13 May 1996.

CLINICOPATHOLOGICAL CONFERENCE

A 34-year-old man withrecurrent melaena afterrenal transplantationAbdul K. Cariem, Mike Arendse, Peter Cruse,

Brian Rayner

Case presentationThe patient a 34-year-old Asian man, was an unemployedprojects engineer whose medical history had begun in 1982when he was diagnosed as having IgA nephropathy on renalbiopsy after an episode of loin pain and haematuria. Initiallyrenal function was normal, but over the ensuing years itgradually deteriorated and by April 1992 he was in end­stage renal failure requiring haemodialysis. During thisperiod he required treatment for moderately severehypertension, and underwent highly selective vagotomy andpyloroplasty for an intractable duodenal ulcer in 1991. Afterthe operation he developed recurrent bouts of waterydiarrhoea. These were episodic, and he passed about 30stools per day. There was no abdominal pain, nausea,vomiting or loss of appetite. He had lost about 3 kg sincesurgery. There were no identifiable precipitating factors. Hismedication was enalapril 10 mg and atenolol 50 mg daily.

On examination the patient had pallor of the nails but nojaundice, lymphadenopathy or features of thyrotoxicosis. Hisabdomen was soft with no masses or visceromegaly. Amidline surgical scar was present. Rectal examination wasnormal and the stool was negative for occult blood. The restof the examination was unremarkable except for milddisplacement of the apex beat and a blood pressure of170/105 mmHg.

The following investigations were done: full blood count ­haemoglobin 8.1 g/dl, mean corpuscular volume 88/1J1, whitecell caunt 7.82 x 10'11, platelets 227 x 10'11, erythrocytesedimentation rate 40 mrn/1 st h, normal differential count;biochemistry - Na- 137 mmol/J, K- 5.5 mmal/J, urea 29.1mmoVl, creatinine 1 026 IJmol/J; B l2 and folate - normal,stool - na red blood cells (RSCs), while blood cells arparasites, cutture negative; fasting gastrin 132 pgIml (normalo- 115 pg/mQ; 72-haur stool - faecal fat normal, mass208 9 (normal < 200 g); upper gastra-intestinal endoscopy- normal; duodenal and rectal biopsies - normal; small­bowel enema - normal; sigmoidoscopy to 20 cm -normal; ultrasound examination of abdomen - normal;thyroid-stimulating hormone - normal; 14c xylose - notsuggestive of bacterial overgrolNth; lactose tolerance -

Clinicopathological conference held in the Department of Medicine,

Groote Schuur Hospital, Cape Town, on 22 June 1995

Clinical Discussant Abdul K. Cariem. M.a. CH.e., F.C.P.

Pathologist Mike Arendse. 8.se. (MED.) HONS. M.se. (MED.). M.e. CI'oLe .• M.MED. (PATH.)

Chairperson: Peter Cruse. 1.1.8. CH.S.. PH.D.• F.R.C. flATH.

Editor. Brian Rayner. M.a. CH.S" F.e.p.. M.MED.

\'olltmt 86 No. 10 October 1996 SAMJ

grossly abnormal H2 breath test; glucose tolerance test ­fasting 5.1 mmoVI, '/2 hour 4.5 mmoVI, 1 hour 6.8 mmoVI,11/2 hours 4.5 mmoVI, 2 hours 4.8 mmoVI.

The patient's symptoms of diarrhoea subsided on alactose-free diet, but he was re-admitted 1 month later withpost-prandial vomiting of coffee-ground material, melaena,epigastric discomfort, fatigue, malaise and dizziness. Onexamination he was markedly pale, there was some epigastrictenderness and the stool was positive for blood. Thehaemoglobin concentration was 5.5 g/dl, and 2 units ofpacked cells were transfused. Upper gastro-intestinalendoscopy was normal and barium meal examination showedsurgical clips at the oesophageal-gastric junction and featuresin keeping with an old duodenal ulcer. An RBC scansuggested duodenal or jejunal blood loss. Gastroscopy wasrepeated with a colonoscope, but this was also normal. Arepeated RBC scan showed no further bleeding 1 week later.Maintenance dialysis was commenced after this episode.

In August 1992 the patient underwent a successfulcadaver renal transplant. He received standard triple-therapyimmunosuppressive treatment (steroids, azathioprine andcyciosporin), with cyclosporin being withdrawn at 6 months.At the time of surgery possible nodes were seen in the rightiliac fossa; biopsy revealed extensive sinus histiocytosis. InNovember 1992 investigations for headache, macroscopichaematuria and fever were done. The macroscopichaematuria was attributed to recurrent IgA nephropathy inthe transplanted kidney, and the headache and fever settledwithout specific treatment.

Between August 1994 and April 1995 the patient had sixepisodes of melaena requiring admission to hospital andtransfusion. There was no obvious precipitating event andhe only experienced symptoms related to blood loss. Themelaena usually subsided after a few days. He had somedyspeptic symptoms which usually settled with symptomatictreatment with antacids. There was no loss of weight, loss ofappetite, abdominal pain, or change of bowel habit exceptduring periods of melaena. A chest radiograph done inDecember 1994 showed a nodular infiltrate in the rightupper lobe, but the patient had no cough, sputum, fever ornight sweats and this was not investigated further. Hismedication was prednisone 7.5 mg daily, azathioprine 100mg daily and several antihypertensive drugs. Renal functionhad been declining progressively owing to chronic rejection,and by March 1995 he once more required haemodialysis.Haematological investigation showed no bleeding diathesis,liver function tests were normal, and cytomegalovirus titreswere always IgG-positive and IgM-negative. The followinggastro-intestinal investigations were all normal: gastroscopy(4 times), colonoscopy (twice), sigmoidoscopy, andangiography (twice). An RBC-Iabelled scan suggested bloodloss from the terminal ileum during March 1995.

Differential diagnosisOr Abdul Cariem: The case under discussion is that of ayoung man who, after renal transplantation, was admitted onsix occasions over an 8-month period with episodes of life­threatening gastro-intestinal haemorrhage, presenting asepisodes of melaena. None of these episodes wereaccompanied by haematemesis, and in addition 4

SAMJI

ARTICLES

gastroscopies were reported as normal. We are not informedwhether or not altered blood was present in the stomach,which, in the absence of a mucosal lesion, would have raisedthe possibility of haematobilia or bleeding from the pancreas.However, the patient does not have the medical backgroundto place him at risk of these complications, i.e. a history ofblunt or penetrating injury to the liver or pancreatitis. Withrepeatedly normal upper endoscopy (including one performedusing a colonoscope, though we are not informed howsuccessful this was and to what extent the proximal smallbowel was examined), 2 negative colonoscopies and 2unhelpful abdominal angiographies, it would therefore appearthat the source of bleeding is localised in the mid- or distalsmall bowel. This conclusion is supported by a labelled RBCscan which suggested blood loss from the terminal ileum.

Routine investigations having been negative, the patientcan be labelled as having gastro-intestinal bleeding ofobscure origin. Small-bowel lesions described' as presentingin this manner are outlined in Table l.

Table I. SmaJl·bowel lesions causing gastro-intestinal bleeding ofobscure origin

Meckel's diverticulumEnteric duplication cystVascular anomalies - arteriovenous malformations,

haemangiomasVasculitisNeoplasmsZollinger-Ellison syndromeCrohn's diseaseTuberculosisSyphilisTyphoid feverHistoplasmosisMesenteric ischaemiaCoeliac diseaseChronic ulcerative jejunitisDrugs (Slow-K, 6-mercaptopurine)

In a series of 131 patients reported by Thompson et al.,2with obscure gastro-intestinal blood loss, 106 of whom wereassessed, the modes of presentation were melaena in 55patients, anaemia in 35, rectal bleeding in 34, haematemesisin 6, and bleeding from an ileostomy in 1. In patients whohad a small-bowel source for the blood loss vascularanomalies were the commonest lesions (16 patients),followed by Meckel's diverticula (9) and leiomyomas (7). Theremaining 9 patients with small-bowel lesions had, inaddition to angiodysplasia, Meckel's diverticulum (4 patients)and with vascular anomaly, ulcerated duodenal diverticulum,solitary jejunal ulcer, chronic pancreatitis, and Crohn'sdisease (1 each). Unless active bleeding is observed from asite where more than one lesion is present, it may be difficultto decide which is the source of blood loss. Let us now lookat this patient's medical background for clues as to thenature of the. distal small-bowel lesion.

He first came to medical attention in 1982 at the age of 21years when he presented with an episode of loin pain andhaematuria. The diagnosis of IgA nephropathy wasconfirmed on renal biopsy. Renal function deterioratedgradually over the next 10 years, and by 1992 he required

•

SAMJ Volume 86 No. 10 October 1996 1285

haemodialysis. His case is typical of IgA nephropathy interms of age of presentation and sex. IgA nephropathy maypresent as asymptomatic microscopic haematuria or asmacroscopic haematuria. It is characterised by mesangialdeposition of 19A immunoglobulins, and in this way bearssimilarities to Henoch-Schonlein purpura. This lattercondition is characterised by muttisystem involvement, andabdominal pain and gastro-intestinal blood loss is welldescribed as a manifestation.3 Atthough deposits of 19Ahave been demonstrated in skin capillaries in patients withIgA nephropathy, it does not involve any other organ system,and the distal ileal lesion cannot be related to this disease.

In 1991 the patient underwent highly selective vagotomyand pyloroplasty for intractable duodenal ulcer disease. Nodetails are given regarding the medical management of thepeptic ulcer disease and the reasons for resorting tosurgery; also, a drainage procedure is not usually requiredwith a highly selective vagotomy. A short while later hepresented with diarrhoea and after a series of investigations,mostly negative exceJ:?t for a grossly abnormal lactosetolerance test, was found to have lactose intolerance. Theprevalence of lactase deficiency is high in adults, particularlythose of Asian and African descent.· The rapid gastricemptying following the surgery resulted in an overload of theavailable lactase and secondary lactose intolerance. Thediarrhoea responded to a lactose-free diet.

One month later the patient presented with coffee-groundvomitus, melaena, and symptomatic anaemia; thehaemoglobin concentration had fallen to 5.5 g/dl. Upperendoscopy was normal and barium meal examination merelyshowed evidence of scarring of old duodenal ulcer disease.A labelled RBC scan suggested duodenal or jejunal bloodloss, but a lesion could not be demonstrated. The incidenceof recurrence of peptic ulcer disease is higher after surgeryfor duodenal ulcers than after surgery for gastric ulcers, andwith a highly selective vagotomy there is a 7 - 10% relapserate.!! There is also a lower incidence of pain and a higherincidence of haemorrhage as the presenting symptom. It isalso possible to miss a small lesion, both on endoscopy andon barium study, in a patient with a scarred and distortedduodenum. None of the patient's subsequent presentationswith melaena were associated with coffee-ground vomitus,and with 4 normal upper endoscopies and presumably noevidence of blood in the stomach it is very unlikely that theepisodes of melaena were due to recurrent duodenal ulcerdisease.

The next possibly relevant factor we need to consider isthat of renal failure and renal transplantation. The patienthad a renal transplant in August 1992 with evidencesuggesting recurrence of the IgA nephropathy in thetransplanted kidney 3 months later. An increased incidenceof occult gastro-intestinal blood loss has been demonstratedin patients with impaired renal function and in those onhaemodialysis. This has been documented at 0.83 ml/day innormal controls, 3.15 mVday in uraemic patients (thought tobe due to impaired platelet function), and 6.25 ml/day Inpatients on haemodialysis (with the use of anticoagulants anadded factor).5 Although we have excluded peptic ulcerdisease as the cause of our patient's episodes of melaena, itmay be useful to note that, contrary to popular perception,recent studies where endoscopy was used as the mode of

investigation, showed the prevalence of peptic ulcer diseasein patients with renal failure to be the same as that in thegeneral population.1 An increased prevalence has beennoted in transplant patients, particularly during the first 5months after transplantation, with a higher incidence ofduodenal ulcers and a lower incidence of gastric ulcers intransplant patients compared with patients on dialysis. Apoor correlation with symptoms has also been noted:!!

Of greater relevance in this patient is the associationbetween uraemia and gastro-intestinal telangiectasia.Results are conflicting, some studies reporting a lowprevalence of gastro-intestinal telangiectasia in uraemicpatients with gastro-intestinal haemorrhage, while othersreport a 24 - 32% incidence of vascular malformations inuraemic patients with gastro-intestinal haemorrhage.lt

•lo

Conversely, in a group of 65 patients with upper or lower gutangiodysplasia, 35% had renal insufficiency.ll In the lattergroup 65.2% of the lesions were found proximal to theligament of Treitz. In the patient under discussion, failure todemonstrate angiodysplastic lesions on upper and lowerendoscopy makes this an unlikely cause of his bouts ofmelaena, though theoretically it is possible for the lesions tobe localised to the area of the small bowel not visualised.Another confounding problem is whether the finding ofangiodysplasia on endoscopy means that it is the cause ofbleeding. In one study 22.5% of patients with right-sidedcolonic angiodysplasia were subsequently found to bebleeding from a small-bowellesion.12

This patient had his first episode of melaena almost 2years after renal transplantation. He was onimmunosuppressive therapy in the form of steroids andazathioprine, having received cyclosporin during the first 6months only. He was admitted with episodes of melaena on6 occasions over an a-month period, and towards the latterpart of that period haemodialysis was recommenced owingto recurrence of IgA nephropathy in the transplanted kidney.We are not told at what stage immunosuppressive therapywas discontinued. Nevertheless, there is a 3 - 10 timesincreased risk of de novo malignant tumours after renaltransplantation. This is seen both in patients who receivedonly steroids and azathioprine and in those receiving themore recent cyclosporin-based immunosuppressiveregimens. These tumours appear to present earlier inpatients receiving cyclosporin, means of 36 and 36.2months after transplantation versus 95 and 75 months,respectively, as has been reported by two groups'3.Urecently. In reports from Germany and Spain'J·,. the skinappears to be the most common site (16 of 31 cases and 11of 26 cases, respectively), whereas a report from Japan1S

found the gut and liver most commonly involved (5 and 6cases, respectively, of a total of 22). The incidence ofmalignant lymphoma was found to be low in all three series- 2 out of a total of 79 cases. Again, although theoreticallyit is possible for our patient's melaena to be caused by amalignant lesion in the distal small bowel, in view of theshort duration of his immunosuppressive therapy we shouldconsider other possibilities.

The last piece of information we are given is a possibleabnormality on the chest radiograph midway through thepatient's 6 admissions with melaena. On review of the relevantradiographs a nodular infittrate in the right upper zone

1286 Volume' 86 No. 10 Oerober 1996 SAMJ

adjacent to the mediastinum is clearly evident. There is noevidence of any breakdown, and no hilar lymphadenopathy.

Opportunistic pneumonias after renal transplantation tendto occur in the first 1 - 6 months after transplantation. Theorganisms involved include cytomegalovirus, Pneumocystispneumoniae, fungi such as Aspergillus, Cryptococcus andCandida, and Gram-negative bacilli such as Usteria andNocardia.'6 The characteristics of the chest radiograph are avital clue to the aetiological agent. In this patient - 2 yearsafter renal transplantation, immunocompromised on thebasis of immunosuppressive therapy as well as impairedrenal function, with a focal nodular infiltrate notcharacteristic of the above infective agents, and noassociated respiratory symptoms - tuberculosis must beconsidered the most likely cause of the infiltrate seen on thechest radiograph, particularly bearing in mind the high localprevalence of this disease.

An increased incidence of tuberculosis in patients withrenal transplants due to uraemia and the use ofimmunosuppressants has been reported worldwide. Theincidence varies with the local prevalence of the disease,ranging from 0.5 - 1% in North America and 1.0 - 4% inEurope to as high as 9.5% in India. l1 Locally, in an 11-year(1980 - 1990) review, 20 cases of tuberculosis occurred in407 transplant recipients available for review, an incidenceof about 5%.'8 The mean time from transplant to diagnosisof tuberculosis was 15 months (range 2 - 74 months).Consolidation on the chest radiograph was present in 13patients, a miliary pattern in 4, pleural effusion in 3,cavitation in 1, hilar lymphadenopathy in 1, and atuberculoma in 1 patient. Extrapulmonary tuberculosisoccurred in 6 patients; 1 had meningitis, 1 renalinvolvement, and 4 disseminated tuberculosis.

Two 10-year reviews of gastro-intestinal tuberculosis havebeen reported from this hospital (1962 - 1971 and 1972 ­1981), a total of 242 patients (117 and 125 respectively).""A total of 95 patients (49 and 46 respectively) had intestinaltuberculosis, 129 tuberculous peritonitis, and 18 mesentericnode involvement. The commonest symptoms experiencedby the patients with intestinal tuberculosis were weight loss(49/49 and 38/46), abdominal pain (43/49 and 37/46),diarrhoea (26/49 and 22/46) and vomitin9 (10/49 and 23/46).The chest radiograph was abnormal in 20/49 and 23/46patients. In the second review,20 of the 46 patients withintestinal involvement, 13 had small-bowel involvement.Five patients presented with malabsorption, of whom 2 hadactive pulmonary tuberculosis. Of the remaining 8 patients,4 presented with obstruction, 2 with perforation, 1 withmelaena, and 1 with pyrexia of unknown origin. The lesionwas localised to the distal small bowel in 7 of these 8patients, and 2 had active pulmonary tuberculosis.

Occult bleeding due to gastro-intestinal tuberculosis hasbeen described in up to 39% of patients.21 In the above­mentioned series of 46 patients with intestinal tuberculosis,2014 had evidence of gastro-intestinal blood loss; this wasoccult in 13 cases, while 1 patient had melaena. However, inanother 2 series describing 109 and 196 patients withgastro-intestinal tuberculosis,22.23 there was no evidence ofgastro-intestinal blood loss, though anaemia was commonand occurred in up to 60% of patients.

Although overt gastro-intestinal bleeding due to intestinaltuberculosis may be rare, a number of isolated case reports

SAMJARTICLES

have appeared describing massive life-threatening gastro­intestinal bleeding. This has complicated lesions in thestomach, small bowel and colon.2

'<-28

In summary, we have a young man post-renaltransplantation on immunosuppressive therapy and Withrenal impairment, with a pulmonary infiltrate consistent withpulmonary tuberculosis, presenting with repeated episodesof massive intestinal bleeding, localised to the distal smallbowel on a labelled RBC scan. Although it is possible thathe may have a totally unrelated intestinal lesion such as aMeckel's diverticulum, or an isolated small-bowelangiodysplastic lesion related to his renal impairment, or amalignant lesion related to his immunosuppressive therapy, Iconsider that the pulmonary infiltrate and the site of theintestinal lesion would favour intestinal tuberculosis as themost likely cause of the lesion and the blood loss. It isdifficult to explain why the infection failed to becomerampant over the a-month period during which the Patientexperienced the episodes of melaena, in view of hisimmunocompromised state. I suspect that theimmunosuppressive therapy was discontinued when itbecame evident that failure of the transplant kidney Wasinevitable, and not continued to the stage at whichhaemodialysis was recommenced.

The diagnostic procedure indicated, in view of the life­threatening nature of the bleeding episodes and positivelocalisation of the lesion with the labelled RBC scan, wouldbe a laparotomy with a view to resection of the involVedsegment of bowel or biopsy to confirm the diagnosis,followed by medical therapy.

Or Abdul eariem's diagnosis. Ileal tuberculosis.

Operative findingsA diagnostic laparotomy was performed. At surgery therewas found to be blood in the small bowel with multiPlenodular lesions fr.om ileum to jejenum. Mesenteric nOdeswere palpable. A small segmental resection of the ileum wasperformed.

Pathological findingsTwo specimens were submitted. One consisted of a lymphnode, 14 x 8 x 4 mm in its greatest dimensions. Cross­section of the lymph node showed caseous necrosis. Tneother was a full-thickness bowel biopsy specimen, 25 mm inlength and 50 mm in circumference. A transverse ulCer waspresent, measuring 20 mm in diameter and 3 mm in length.

MicroscopySections taken from the ulcer show caseous granulOmatousinflammation replete with Langerhans-type giant cells in thebase of the ulcer. The inflammation extends transmUrallywith interruption of the external muscular layer. Sectionstaken adjacent to the ulcer show caseous granulomatousinflammation within the submucosa of the specimen (Fig. 1).

Sections of the mesenteric node show caseousgranulomatous inflammation replacing the normal nOdalarchitecture. Stains for acid- and alcohol-fast bacilli revealedstainable bacilli within Langerhans-type giant cells.

SAMJ Volume 86 No. 10 OCtober 1996 1287

Fig. 1. A full-thickness biopsy of the terminal ileum showsfeatures of granulomatous inflammation at the base of the ulcer.

Anatomical diagnosis. M. tuberculosis infection of thesmall bowel.

REFERENCES

1. Spechler SJ. Schimmel EM. Gastrointestinal tract bleeding 01 unknown origin. ArchIntern Med 1982: 142: 236-240.

2. Thompson IN, saJem AA. Hemingway AP. et al. Specialist Investigation of obscuregastrointestinal bleeding. Gut 1987; 28: 47-51.

3. Glassock FU. Brenner BM. The major gJomerulopathies. In: Wilson JO. Braunwald E.Isselbacher KJ. et aJ.. eds. Harrison's Principles of Internal Medicine. 12th 00. NewYor1c McGraw-Hill, 1991: 117Q-11SQ.

4. Bllller HA. Grand RJ. Lactose intolerance. Am Rev Med 1990; 14: 141.5. Mathews JB, Silerl W. Operations for peptic ulCef" disease and earty postoperative

complications. In: Sleiseng8l" MH, Fontuan JS, eds. Gasuoinrestina/ Disease.ParhophysK)iogy. Diagnosis, Management. 5th ed. Philadelphia; WB Saunders, 1993:713-730.

6. Aosenblan SG, Drake S, Fadem S, W6ch A, l.ifschitz MD. Gasuointestinal blood lossitl patients wfth chronic renal failure. Am J Kidney Di$ 1982; 1: 232-236.

7. Bemersen B. .Johnsen A, Straume B. BorhoI PG. Jenssen TO, Stakkevold PA­Towards a true prevaleoce of peptic ulcer; the Soneisa gastI'OitttestinaJ disordersD.ldy. Gut 1990; 31: 989-992_

8. Musola R. Franzin G, Monl A. Manfrini C. Prevaience of gasuointestinaJ lesions inurerric pa!ients undergoing dialysis and after renal transplantation.. GaszroinresrEndcsc 1904: 30; 343-346_

9. Zuckennan GR, Cornene GL. CIouse RG, Hanel" HA. Uppet" gastn:Jintestinal bleedingitt patients with chronic renal fallure. Ann Intern Med 1985: 102: 588-592-

10. MatCuard SP, Weinstock.N. Gasuointestinalang~in renal failure. J CIinGastroenterol1988; 10: 482-434.

11. Nawab F. MasU!'S P, SlJbtamani R, Ortego TJ, Thompson CH. Angiodysp/asia inpatients with renal insufficiency. Am J Gastmetlterol 1989; 84: 1297-1301.

12. Steger AC, GaUand AB, Hemingway A, er al. Gastrointestinal haemonhage from asecond C3USl! itt patients with colonic angiodysplasia Br J $urg 1987; 74: 726-727.

13. Riners B, Grabensee 8. Heering P. Malignancy under immunosuppressive therapyincluding cyclospolitte. TfIIlsplant Proc 1994: 26: 2656-2657.

14. VilardeU J. Oppenheimer F, TeJbot-Wright R. er al. Increased risk of malignant turnersin renal transplant recipients receiving cyclospo<ine. Transplant Proc 1992; 24: 1948.

15. Suzuki S, Tanaka K, Ohsaka Y. et al. Development of de IWI'O malignancies followingrenal transplantatiott: a single centre srudy. Transpianr Proc 1994: 26: 938-940.

16. McCabe RE. Diagnosis of pulmonary infections in immunocompromised patients.Med Clin North Am 1988: 72; 1067-10~.

17. KoseIj M. ButIJtl)vW: J, Malovrh M. Tuberculosis in renaJ eJlogtaft recipients.Transplant Proc 1992; 24: 1909-1910.

18. Hall CM, Willalll PA, SWanepoel CA. Mycobacterial ittfectiOtt in renaJ transplalltrecipients. Presentatiott at the Annual Research Day. Department of Medicine.Urtiversity of Cape Towtt. 17 Sep 1992.

19. Novis BH, Bank S, Marils IN_ GastrointesMal and peritoneal luben;ulosls: a study 01cases at the Groote Schuur Hospital. 1962-1971. S AIr Med J 1973: 47: 385--372-

20. GiIinsky NH, Marb IN, Kottlel'RE. Price SK. Abdominal tuberculosis: a IQ-yearreYiew. S Air MedJ 198:3; 64: 849-857.

21. Homan WP, Grate WR, Oitteett P. A «-year experience WIth tubercub.ts entenx:oli1is.World J SuTg 19n; 1: 245-250.

22. I<Iirnadi OE. Qrmerod LP. Gastrointestinal tuberculosis: a retrospective review of 109cases itI a district geoeraI hospital. 0 J Med 1985; 56; 569-578.

23.. Bhansali SK. Abdominal lUbefeuIc:lsls: experience with 300 cases. Am J Gastroenrerof19n; 61: 324.-337.

24. Weissman 0, Gumaste W, Dave PS, Keh W. Bleeding from a tubereulous gastriculcer. Am J Gasrroenteml1990; 85: 742-744.

25. Rais N, Plumber ST. Undre AA,B~SO. Masslve lower ga5tn:ointestittalhaett'lOlThage as a complicatiott 01 intestinal tuberculosis. J Assoc Physicians India1987; 35: 647--648.

26. Waghrnare 6G, Holay MO. Cas RN, Kher A. Massive rectal bleeding due to colonictuberculosis. J Assoc Ptlysicjans India. 1988: 36(6): 392·393.

27. Pozniak AL. Dalton-Clark HJ, Ralphs ON. Colonic tuberculosis presenting as massiverectal bleeding. Tut:JercJe 19a5; 66: 295-299.

28. Wainsztein N, Roe! J, Morales JC. [Intestinal h8fT1OrrtJage due to tuberculosis in apatient with a kidney U"&nsplant] [Spanish]. MedidruJ (8 Aires) 1985; 45(6): 663-666.

Accepted 1 May 1996.

SHORT REPORT

An analysis of DTP­associated reactions bymanufacturer, batch,vaccinator, series numberand infant weightJ.G.Benade

Objectives. To detennine whether two commonly usedDTP batches manufactured by Rh6ne-Poulenc Rorer weremore reactogenic than two commonly used batches

manufactured by the South African Institute of MedicalResearch.

Design. Prospective study.Setting. Six community clinics.

Patients. Infants routinely scheduled for their first threeDTP immunisations.

Main outcome measures. Local and systemic adverse

reactions following immunisation with DTP.Results. Local reactions were significantly more

common with both Rh6ne--Poulenc Rarer products.

Conclusion. AJI adverse reaction rates comparedfavourably with those reported by the Centers for DiseaseControl.

S Atr Med J 1996; 86: 1288-1290.

Because of inadequate surveillance systems in many localauthorities, it is impossible to substantiate a growingperception that the diphtheria-tetanus-pertussis vaccine,DTP-Merieux, manufactured by Rh6ne-Poulenc Rorer, ismore reactogenic than DTP manufactured by the SouthAfrican Institute for Medical Research (SAIMR). Since thisclimate of uncertainty can eventually undermine theExpanded Programme on Immunisation, the validity of theseperceptions was investigated by a stUdy of the reactionprofiles of four commonly used vaccine batches, suspectedof unacceptable reactogenicity by some local authorities.

LiteratureIt is well known that whole-cell pertussis--containing vaccineis more reactogenic than most of the vaccines routinetyused for immunisations.U The reported incidence oftenderness, erythema, swelling or induration at the injectionsite varies from 300 - 700/1 000 DTP doses, USUally occurswithin 48 hours of vaccination and is mostly setf-limited.l

-6 Anodule may occasionally be palpable at the injection site for

Department of Community Health, University of Stellenbosch,Tygerberg, W. Cape

J. G. Benade. M.a OlB.• M.MEO. (C.H.)

Volume 86 No, 10 Ocrobu 1996 SAMJ