Plausible Diagnosis Research Articles

Calcified cysts in the upper abdomen

A plain chest radiograph reveals multiple calcified cysts in the upper abdomen of a man who has been on dialysis for 15 years. What is the most plausible diagnosis?

A case of round ligament varicosities during pregnancy

Estimated incidence of round ligament varicosities in pregnancy is not known and often times it is confused with inguinal hernia due to their clinical similarities. When a patient is presented with inguinal mass especially in association with varicosity in the genital region or lower extremity, round ligament varicosity must be considered as a plausible diagnosis. Depiction of “bag of worms” on color Doppler ultrasonography is diagnostic of the round ligament varicosity and it is known to resolve spontaneously following delivery. We report a case of round ligament varicosities that was diagnosed at 29 weeks of gestation with a brief review of the literatures.

Explicit Reasoning, Confirmation Bias, and Illusory Transactive Memory

Teamwork is important in medicine, and this includes team-based diagnoses. The influence of communication on diagnostic accuracy in an ambiguous situation was investigated in an emergency medical simulation. The situation was ambiguous in that some of the patient's symptoms suggested a wrong diagnosis. Of 20 groups of physicians, 6 diagnosed the patient, 8 diagnosed with help, and 6 missed the diagnosis. Based on models of decision making, we hypothesized that accurate diagnosis is more likely if groups (a) consider more information, (b) display more explicit reasoning, and (c) talk to the room. The latter two hypotheses were supported. Additional analyses revealed that physicians often failed to report pivotal information after reading in the patient chart. This behavior suggested to the group that the chart contained no critical information. Corresponding to a transactive memory process, this process results in what we call illusory transactive memory. The plausible but incorrect diagnosis implied that the two lungs should sound differently. Despite objectively identical sounds, some physicians did hear a difference, indicating confirmation bias. Training physicians in explicit reasoning could enhance diagnostic accuracy.

Malignant Primary Pulmonary Tumor with Hemangiopericytoma-Like Features: Conventional Hemangiopericytoma Versus Solitary Fibrous Tumor

Although extremely rare, the primitive lung hemangiopericytoma in adults is similar to hemangiopericytomas with other soft tissue localizations. Although generally benign and curable after radical surgery, it might also have a malignant clinical course with dissemination in both lungs, infiltration of vital organs (heart, pulmonary artery), extension to the adjacent tissues, and even pulmonary metastases. The treatment of choice is the complete tumor resection with negative surgical margins after excision. Certain histologic features might indicate a malignant potential. The clinical outcome of patients is variable: some are cured after radical surgery and others might present relapse and recurrences that necessitate a second intervention, radiation therapy, and/or chemotherapy. Over the years, the conventionally-defined hemangiopericytoma concept has evolved because of the nonspecific histologic growth pattern (characteristic monotonous appearance, moderate or high cellularity, and a well-developed branching vascular pattern) shared by numerous, unrelated benign or malignant lesions. Hemangiopericytoma is no longer considered a specific entity but rather as a characteristic growth pattern; in the heterogeneous group of hemangiopericytoma-like neoplasms, many lesion categories have been individualized and defined. We report an uncommon case of primitive lung tumor exhibiting hemangiopericytoma-like features, with an aggressive, fatal clinical course. Because of the major histologic overlap between solitary fibrous tumor and hemangiopericytoma and lack of clear classification criteria, we encountered difficulty in including this case in a known clinical entity; primitive solitary fibrous tumor of the lung, which mimics lung hemangiopericytoma, seemed to be the most plausible diagnosis. We discuss the case particularities and the radiologic and pathologic correlations.

Severe spinal cord ischemia subsequent to fibrocartilaginous embolism

Fibrocartilaginous embolism is a rare cause of spinal cord ischemia. Here we report the case of a young previously healthy man who noted sudden thoracic spinal belt-like pain after intensive physical effort. Following a free interval he developed paraplegia, complete sensory loss below Th(4) and inability to voluntarily purge bladder and bowel. Neuroimaging exposed an intramedullary longitudinal hyperintense signal from C(6) down to the conus in T2-weighted images, intersomatic disc collapses and vertebral body infarctions (C(5-7)/Th(8-10)). Other plausible diagnosis, e.g. spinal contusion, cord compression or acute onset transverse myelitis were excluded. Altogether, clinical presentation, neuroimaging and lack of evidence of other plausible diagnosis suggest fibrocartilaginous embolism as the most probable diagnosis.

A case study of possible differential diagnoses of a medieval skeleton from Denmark: leprosy, ergotism, treponematosis, sarcoidosis or smallpox?

AbstractThis paper uses macroscopic and radiological examinations to provide differential diagnoses of pathological lesions in the skeleton of a young woman, 20–25 years of age, which triggered the Danish palaeopathologist Vilhelm Møller‐Christensen's interest in leprosy. The skeleton was incomplete, but the majority of bones of the upper body, as well as the skull, were present. The pathological changes consisted of medullary and cortical lytic foci, periosteal reaction and enhanced cortical density. The lesions were most extensive on the left side, especially around the elbow, wrist and scapula. Treponematosis, leprosy, smallpox, ergotism, rheumatoid arthritis, tuberculosis and sarcoidosis are all reviewed with regard to bone and joint pathology and their likelihood of being the correct diagnosis. We concluded that the most plausible diagnosis is treponematosis, but neither sarcoidosis nor smallpox can be completely excluded. Copyright © 2007 John Wiley & Sons, Ltd.

Central nervous system graft-versus-host disease: report of two cases and literature review

Two patients with a plausible diagnosis of central nervous system graft-versus-host disease (CNS-GVHD) are described. Both presented with neurological manifestations 6 and 18 months following allogeneic transplant with hemiparesis, seizure, encephalopathy and magnetic resonance findings of hyperintense white matter lesions on T-2 weighed images. Brain biopsy in one and autopsy in the other revealed profound perivascular lymphocytic infiltrates composed predominantly of T-lymphocytes that were of donor origin. Although an unequivocal diagnosis of CNS-GVHD is difficult to establish, the transplantation community should be aware of this controversial entity.

Does Your Child Have Asthma? Filled Prescriptions and Household Report of Child Asthma

Introduction Surveys are central for information on asthma prevalence. Recently, the validity of parental reports of pediatric asthma has been questioned. Confidence is examined in the report of asthma for children, obtained in a survey from the adult household member most knowledgeable about household health care (MKA). Method MKA reports of asthma are compared with pharmacy records of prescriptions beneficial in asthma treatment (“asthma medications”) for children 0 to 17 years old in the 1996 Medical Expenditure Panel Survey. Results “Asthma medications” were filled for 6.5% of children, yet the MKA did not report asthma for 47.3% of them. However, for 61.2% of these children, the MKA reported plausible alternative medical conditions. For 9.0%, diagnosis information was missing. Among children with an “asthma medication,” the MKA was less likely to report either asthma or a plausible alternative diagnosis for girls and for children 0 to 5 years of age. Reporting was not statistically different by child race/ethnicity, household income, education level, and MKA English language proficiency. Discussion Surveys do not overlook as many children with asthma as previously reported. Among children with “asthma medications,” only sex and age appear to be different for children whose MKA reported either asthma or a plausible alternative diagnosis versus those whose MKA did not report either.

Familial Adenomatous Polyposis and Factor VII Deficiency in a Young Man

To report a case of Familial adenomatous polyposis in a young man with concomitant Factor VII deficiency. A 29 y/o African American male presented with complaints of weakness and recent shortness of breath. He attributed the symptoms to several nosebleeds in the past six months. He denied any trauma or other medical history. Patient reported frequent nose bleeds for last six months though denied any vomiting, malena, hematochezia or other sources of bleeding. His father died at age of 42 with colon cancer and uncle also had colon cancer but denied any bleeding disorders in family. He was a smoker, occasional alcohol user but denied illicit drug abuse. On examination, he was tachycardic with blood pressure of 110/50. Rest of the exam was remarkable only for heme positive brown stool on rectal exam. Lab data showed hematocrit 11 with MCV 67, Platelet 473K, TIBC 431, Serum Iron 9 and Ferritin 3 ng/mL. His coagulation profile showed PTT elevated to 20.7 seconds and INR elevated to 1.8. He was admitted and received 3 units of packed red blood cells. With profound iron deficiency anemia and strong familial history of colon cancer, gastrointestinal consultation was sought. Panendoscopy was remarkable for innumerable polyps in colon. FAP was most plausible diagnosis. Biopsies revealed adenomatous polyps without high-grade dysplasia or malignancy. Coagulation parameters did not correct after transfusions. Mixing study showed correction of prothrombin time. Coagulation factor deficiency seemed possible cause. Factor VIII and IX were normal. Factor VII levels was drawn later, after hematology consultation, it was 12 (reference 70–130). A diagnosis of factor VII deficiency was made. He underwent colectomy and received recombinant factor VII perioperatively. Patient was counseled and genetic counseling was advised along with surveillance in other family members. This is an interesting presentation of two rare familial disorders in same patient. Familial adenomatous polyposis is autosomal dominant diseases caused by mutations in the adenomatous polyposis coli on chromosome 5q21. FAP occurs in approximately 1/10000–30000 live births, and accounts for less than 1 percent of the total colon cancer risk in the USA. With estimated prevalence of 1: 500 000, inherited FVII deficiency is also a rare, autosomal recessive coagulation disorders. The FVII gene (F7) is located on chromosome 13. There has been no reported link between these two relatively rare disorders and this might be just a coincidence or not!

Two sisters with IMAGe syndrome: Cytomegalic adrenal histopathology, support for autosomal recessive inheritance and literature review

Adrenal hypoplasia congenita (AHC) is a rare condition and causes primary adrenal insufficiency. X-linked (OMIM 300200) and autosomal recessive (OMIM 240200) forms are recognized. Recently, an association between Intrauterine growth restriction, Metaphyseal dysplasia, Adrenal hypoplasia congenita, and Genital abnormalities (IMAGe syndrome; OMIM 300290) has been described. We present the clinical features of two sisters with intrauterine growth restriction, AHC, and dysmorphic features. Interesting histopathologic findings of one sister are also presented. We suggest that IMAGe syndrome is the most plausible diagnosis and that autosomal recessive inheritance is likely. We analyzed genes that were postulated candidates for IMAGe syndrome (SF1, DAX-1, and STAR), and no mutations were found. Other cases of IMAGe syndrome are reviewed.

Spondyloarthropathy presenting at a young age: case report and review

The diagnosis of juvenile spondyloarthritis (JSA) is rarely entertained in young children who present with back and leg pain. We present a case of a 6-year-old male who presented with a 3-year history of severe back and leg pain and a positive Gower's sign, and was given a presumed diagnosis of muscular dystrophy. Presenting serologic evaluation included a mildly elevated sedimentation rate and C-reactive protein (CRP). Computed tomography of the pelvis demonstrated large erosions affecting both sacro-iliac joints. Despite the unusually young age of this patient, ankylosing spondylitis seemed the most plausible diagnosis. Following rheumatological evaluation and treatment for JSA, he showed significant clinical improvement. His disease, however, has not entirely remitted with signs of enthesitis at the Achilles tendon and knees. We present this case to illustrate that JSA could account for symptoms at an early age and not considering it could lead to multiple medical visits and diagnoses. To our knowledge, based on a search of the World literature, this would appear to be the youngest case of JSA reported with demonstrable severe sacroiliitis.

Rheumatoid arthritis, Pyrexia of Unknown Origin (PUO) and pancytopenia

A 61-yr-old previously fit lorry driver was referred by his general practitioner in October 1999 to an orthopaedic clinic with a 6-month history of painful hands, shoulders and feet. His general practitioner had treated him with ibuprofen without much improvement. He admitted to a recent history of loss of appetite with loss of two stones (12.7 kg) in weight. His joint pains had been increasing in severity with visible swelling of his fingers and knees. He had no relevant family history. He was an ex-smoker and drank little alcohol. Physical examination revealed a healthy-looking man with signs of synovitis in the symptomatic joints. The orthopaedic surgeon referred him to the rheumatologist (K.C.). The preliminary blood tests showed haemoglobin of 10.4 g/dl with normal haematocrit, leucocyte and platelet counts and normal serum vitamin B12 and red-cell folate levels. His serum ferritin was 468 g/l (normal range 18–400 g/l). He also had transient random hyperglycaemia, which was followed by a normal glucose tolerance test. His rheumatoid factor was elevated at 1:2560. His chest X-ray was normal except for some thickening of the bronchiolar wall. He was seen in the rheumatology clinic 2 months later, in December 1999, with classical symptoms and signs of a symmetrical inflammatory polyarthritis suggestive of rheumatoid disease. His knees were drained and injected with methyl prednisolone and local anaesthetic. He was also given intramuscular methyl prednisolone (120mg) to induce remission of active disease. Investigations revealed haemoglobin 10.3 g/dl, white cell count 6.4 10/l, platelet count 449 l0/l and ESR of 100mm in the first hour. The biochemical screens were all normal, including liver, renal and bone panels. The rheumatoid factor was 2460 IU (normal range 30 IU) and CRP was elevated at 85.1mg/dl (normal range <10mg/dl). The X-rays of his hands and feet showed no obvious erosions. The option of commencing methotrexate as a second-line drug was discussed but he preferred to wait until he returned from a planned Christmas holiday in Spain. Soon after his return from holiday he developed sudden severe cramp-like pains associated with numbness in his legs. He was virtually unable to walk. An MRI scan of his whole spine was done on the advice of the neurosurgeons, and, as the scans were normal, he was referred to the neurologists, who performed a lumbar puncture to exclude acute infective polyneuritis. The neurologists also noted a skin rash on his legs associated with left foot drop (Fig. 1). An autoimmune disease was suspected and the rheumatologist (K.C.) was consulted. On examination he had a vasculitic rash with peripheral neuropathy. There was no lymphadenopathy or organomegaly. Given the recent diagnosis of seropositive rheumatoid arthritis (RA), systemic rheumatoid vasculitis (SRV) seemed to be the most plausible diagnosis. Investigations at this time revealed haemoglobin 9.0 g/dl, white cell count 12.1 10/l and neutrophils 9.6 10/l with normal platelet and lymphocyte count. The ESR was 110mm in the first hour and CRP was 121mg/dl (normal range <5mg/dl). The biochemical, serological and microbiological screens including hepatitis A, B and C were all normal/negative. His anti nuclear antibodies, anti DNA antibodies, extractable nuclear antigens, anti cardiolipin antibodies, anti neutrophil cytoplasmic antibodies and cryoglobulins were all negative. Echocardiogram was normal. A cerebrospinal fluid study showed a mild increase in cerebrospinal fluid protein but no oligoclonal band or albumino-cytological dissociation. Electromyography showed severe sensorimotor polyneuropathy of the axonal type with some denervation changes in the peroneal muscles. A skin biopsy confirmed active necrotizing vasculitis, consistent with SRV (Fig. 2). He was commenced on pulse intravenous methyl prednisolone (500mg) and cyclophosphamide (500mg) with some symptomatic improvement, and was discharged home on a 2-weekly regime.

Portfolio assessment using a structured interview

Context and setting The use of portfolios, as a learning and assessment tool, has gained considerable popularity in the past decade. We adopted portfolios as part of our assessment package in the clinical clerkships (years 4–6) of our extensively revised MBChB programme in 2002. Why the idea was necessary A major limitation of portfolios is the resource-intensive nature of the assessment process. Published data report examination times exceeding 90 minutes per candidate. In resource-constrained environments, typical of developing countries, this time requirement is prohibitive. What was done During the fourth-year medicine clerkship, students wrote up 25 patient encounters during a 14-week clerkship and reflected on these by: (a) editing (in another colour ink) their clerking notes after discussion with senior clinicians and review of standard reference texts, and (b) formulating written question and answer tasks regarding some aspect of individual patient encounters. The latter were dictated largely by individual student learning needs with guidance by clinician tutors where necessary. At the end of the year portfolios were examined by interview. Four patient encounters were discussed during a 30-minute single-examiner interview; candidates selected 1 case and examiners the remaining 3 cases. Using 5 structured questions, candidates’ ability to synthesise clinical assessments, using information gathered during patient encounters, was explored. The questions determined whether candidates had (a) defined clearly the patient’s presenting problem; (b) offered a pathophysiologically plausible diagnosis; (c) substantiated the diagnosis using available clinical and investigatory findings; (d) considered a reasonable differential diagnosis; and (e) formulated a reasonable management plan. Responses to questions were rated using a 9-point global rating scale: poor (1–3), adequate (4– 6) and good (7–9). Clinician examiners assigned a final percentage score to each case using a criterionreferenced scale: 3 or more questions rated ‘poor’ scored 45% or less, 2 or fewer questions rated ‘poor’ scored 52–58%, all questions rated ‘adequate’ scored 60–62%, 2 or fewer questions rated ‘good’ scored 65– 68%, at least 3 questions rated ‘good’ scored 70–74% and all questions rated ‘good’ score 75% or more. Clinician examiners were trained before implementing this assessment tool. Pearson’s correlation coefficient, Cronbach’s alpha coefficient and the standard error of measurement (SEM) of the assessment tool were determined using STATISTICA (Statsoft). Evaluation of results and impact The results of the fourth-year class of 2004 (n 1⁄4 181) were evaluated. The mean (± SD, 95% CI) portfolio interview score achieved was 67.5% (± 10.5, 66–69.1). The correlation coefficients for the portfolio interview, when compared to the multiple choice written examination and clinical case-based examination, were, respectively, r 1⁄4 0.42, and r 1⁄4 0.37. Cronbach’s alpha coefficient for the portfolio interview was 0.88 with an inter-item correlation of 0.66. The SEM was 3.6. The single-examiner portfolio interview, using standardised questions and a global rating scale, demonstrates high internal consistency and broadens the spectrum of assessment currently achieved. This method, requiring less examination time per candidate than published data, has been added to the array of assessment tools used in the clinical clerkship rotations in our new programme.

61-Year-Old Woman With Knee Pain and Confusion

61-Year-Old Woman With Knee Pain and Confusion

The contribution of the EEG technologists in the diagnosis of Panayiotopoulos syndrome (susceptibility to early onset benign childhood autonomic seizures)

Purpose: To assess the contribution of the EEG technologists in the diagnosis of children with epileptic seizures. Methods: We analysed the clinical information obtained by the EEG technologists from children with epileptic seizures and their parents, and assessed its value for the generation of a clinically useful EEG report and a plausible electroclinical diagnosis. Interviews were based on a qualitative questionnaire, and were videotaped. We focused on Panayiotopoulos syndrome (PS) because it has a high rate of misdiagnosis, usually for encephalitis or other severe cerebral insults. Results: Between 1998 and 2001, 424 EEG were performed in 308 children aged 1–14 years, of whom 228 (74%) had one or more epileptic seizures. We diagnosed PS in 14 children (6.1%), mainly based on clinical information. Three other had symptomatic ictal vomiting. In 9 of the 14 children with PS, diagnosis was achieved by the information collected by the EEG technologist. Five of these children were being treated for encephalitis, and management was altered accordingly. In a further three children the diagnosis of PS was confirmed. Conclusion: These findings demonstrate that the contribution of the EEG technologists to the diagnosis of people with epilepsies can expand well beyond their established role of recording and describing an EEG. We propose that technologists should be actively involved in prospective electroclinical studies if carefully designed protocols are used.

Comparison of four clinical prediction scores for the diagnosis of lower limb deep venous thrombosis in outpatients

Purpose We compared three scores for the prediction of deep venous thrombosis with a new score designed specifically for outpatients. Methods Patients referred for evaluation because of suspected deep venous thrombosis were examined by ultrasonography. Sensitivity and specificity were calculated for three clinical scores (Wells [nine components], Kahn [four components], and St. André [six components]). We developed a new score by multivariate analysis, and then compared this score with the others in a new sample. Results Four hundred and forty-four outpatients were included in the first sample, of whom 126 (28%) had deep venous thrombosis. The Wells score was a better predictor of deep venous thrombosis than the Kahn and St. André scores. According to the Wells score, 73 patients had a high probability of deep venous thrombosis (of whom 51 [70%] actually had a thrombosis) and 178 had a low probability of deep venous thrombosis (of whom 19 [11%] had a thrombosis). A new score was developed as follows: male sex (+1), lower limb palsy or immobilization (+1), confinement to bed >3 days (+1), lower limb enlargement (+1), unilateral lower limb pain (+1), and other plausible diagnosis (–1). In a validation sample of 282 outpatients, this score identified 31 patients who had a high probability of deep venous thrombosis (score ≥3), of whom 18 (58%) had a thrombosis, and 70 patients who had a low probability (score ≤0), of whom 3 (4%) had a thrombosis. The Wells score and this ambulatory score had similar test operating characteristics in the validation sample. Conclusion Our new six-component score had similar diagnostic utility as the nine-component Wells score among outpatients being evaluated for deep venous thrombosis.

Degree Operators and Scope

n/a

Lymphonodular Hyperplasia as a Sign of Food Allergy in Children

To the Editor: About the time the article “Lymphonodular Hyperplasia as a Sign of Food Allergy in Children” by Kokkonen et al. (J Pediatr Gastroenterol Nutr 1999;29:57–62) was published, we had two cases with similar endoscopic and histologic findings. Our cases involved two girls, one 2 years and one 14 months of age. Both were brought for consultation because of poor weight gain with no other abdominal symptoms. Because they were positive for anti-gliadin and anti-endomysium antibodies, they were scheduled for an endoscopy. Duodenal biopsy specimens of the younger girl showed subtotal villous atrophy, supporting the possibility of celiac disease that will have to be confirmed in the future. The other girl had normal villi. What was surprising was to find lymphonodular hyperplasia in the duodenal bulb of both girls. The visual finding was corroborated by biopsy specimens obtained in the affected area. Although Kokkonen et al. encountered a duodenal bulb frequently (12 out of 63 children), in our series of approximately 140 to 150 endoscopies per year in the past 20 years, it is the first time we have seen such an image. We do not believe we have overlooked this finding in our endoscopies because, when present, it is very evident. The literature on intestinal lymphonodular hyperplasia usually refers to its more frequent location in the colon or ileum (1), but there is little said about the duodenal bulb. This is probably related to a smaller prevalence of these pathologic findings than that stated by the authors. We cannot explain why we had never seen an image similar to these and then in 1 month had two similar cases. Is the incidence growing in our population? We cannot yet answer this question. As for the reason for the lymphonodular hyperplasia, we do not believe food allergy is the only, or even the main, causative agent. Our patients had no signs or symptoms of food allergy. In one girl, celiac disease seems to be a plausible diagnosis. Brocchi et al. (2) have described duodenal bulb micronodules in a patient with celiac disease. We believe duodenal bulb lymphonodular hyperplasia is a pathologic finding that should prompt further studies to diagnose the cause of aggression into the intestinal tissue. A wide range of aggressive agents, including parasites, celiac disease, and an allergic response to various foodstuffs, should be evaluated. Lucrecia Suárez Cortina Amaya Bélanger-Quintana

Long-term outcome of patients with unexplained syncope treated with an electrophysiologic-guided approach in the implantable cardioverter-defibrillator era

OBJECTIVESWe evaluated the long-term outcome of patients with coronary artery disease and unexplained syncope who were treated with an electrophysiologic (EP)-guided approach.BACKGROUNDElectrophysiologic studies are frequently performed to evaluate unexplained syncope in patients with coronary artery disease. Patients with this profile who have inducible ventricular tachycardia are considered at high risk for sudden death and increased overall mortality, and therefore are often treated with an implantable cardioverter-defibrillator (ICD). The impact of this EP-guided strategy is unknown because there are no data comparing the long-term outcome of ICD recipients with that of noninducible patients.METHODSWe evaluated 67 consecutive patients with coronary artery disease and unexplained syncope. All patients were treated with an EP-guided approach that included ICD implantation in patients with inducible ventricular tachycardia.RESULTSElectrophysiologic testing suggested a plausible diagnosis in 32 (48%) of these patients. Inducible monomorphic ventricular tachycardia was the most common abnormality. Despite frequent appropriate therapy with ICDs, the total mortality for patients with inducible monomorphic ventricular tachycardia was significantly higher than for noninducible patients. The respective one- and two-year survival rates were 94% and 84% in noninducible patients and 77% and 45% in inducible patients (p = 0.02).CONCLUSIONSElectrophysiologic testing suggests an etiology for unexplained syncope in approximately 50% of patients and risk stratifies these patients with regard to long-term outcome. Patients who receive an ICD for the management of inducible ventricular tachycardia have a high incidence of spontaneous ventricular arrhythmias requiring ICD therapy. However, despite ICD implantation and frequent appropriate delivery of ICD therapies, patients with inducible ventricular tachycardia have a significantly worse prognosis than do those who are noninducible.

Deaths among women of reproductive age in Cape Verde, causes and avoidability

In 1992-93 all deaths (n=97) of women 15 to 49 years old in three islands of Cape Verde were investigated to determine the cause of death, the maternal mortality ratio, the reliability of cause-of-death ascertainment, and the level of avoidability. Data were obtained through interviews with the deceased person's family members and other knowledgeable persons (verbal autopsy) and through hospital files. The overall mortality rate was 163 per 100,000 women 1549 years old, and the maternal mortality ratio was 127 per 100,000 live births. A plausible diagnosis could be determined in 77%. The most frequent causes of death were circulatory disorders, external causes, maternal causes, infectious diseases. and neoplasms. It was estimated that three of four of the deaths were avoidable with locally available resources. Since access to health care in the study area is not a major hindrance, a further decrease of female mortality depends mainly on improved quality of care in health facilities.