Simple Algorithms for the Management of Genital Ulcers: Evaluation in a Primary Health Care Center in Kigali, Rwanda

Article excerpt

Introduction

Prompt diagnosis and treatment of genital ulcer disease (GUD) is important not only to reduce morbidity but also to slow down the spread of human immunodeficiency virus (HIV) (1). In developing countries, where laboratory facilities are scarce, etiological diagnosis of GUD is usually based on clinical criteria only. However, even before the HIV era, it was reported that this approach was not very accurate, even if performed by experienced clinicians (2, 3). To assure prompt and effective treatment of patients with sexually transmitted diseases (STDs) at the primary health care level, WHO has developed simple flowcharts (4), including two designed for the management of GUD.

In Rwanda, which has a population of around 7 million, about 5000 cases of primary syphilis are reported annually to the health authorities, other causes of GUD are reported only sporadically or not at all. In the Centre Medico Social de Bilyogo in Kigali in 1985, the proportions of chancroid, syphilis and genital herpes diagnosed among patients with GUD were 18%, 28%, and 19%, respectively; 59% of these patients were infected with HIV-1 (5).

The first objective of this study was to assess the proportion of genital herpes, syphilis, and chancroid in patients with and without HIV infection who presented with GUD at a primary health care centre in Kigali. The second, was to compare three simple methods for the management of GUD (two WHO algorithms (flowcharts) and a clinical approach) to determine which approach would result in the largest proportion of patients with chancroid and/or syphilis receiving the correct treatment.

Patients and methods

Data collection

During 1990-92, on three working days each week, all consecutive men and women presenting with genital ulcers at the Centre Medico Social de Bilyogo in Kigali were included in the study. This primary health care centre serves the lower socioeconomic levels and is situated in a part of the city where prostitution is widespread.

Demographic and clinical information about the patients were obtained in a standard interview. All patients underwent a physical examination of the external genitalia and the inguinal region before specimens for laboratory analysis were taken; each patient was diagnosed clinically by a physician (JB) before the laboratory results were known. The following criteria were used: invasive ulcers were considered as chancroid; noninvasive ulcers, as primary syphilis; and genital herpes was diagnosed if vesicles were present, or if there was a history of recurrences, or the ulcers were superficial (erosions). Pain and purulence were not used as diagnostic criteria. If the clinical picture did not correspond to one of these criteria, the diagnosis remained undetermined. No attempts were made to identify clinically other causes of GUD or mixed infections. All patients were requested to return for clinical and microbiological evaluation on days 7, 14, 21 and 28 after the initial visit.

Laboratory procedures

Haemophilus ducreyi was isolated by inoculating a swab specimen from the ulcer directly onto two selective media. The first medium consisted of Mueller-Hinton agar base 2 (BioMerieux, Marcy l'Etoile, France) supplemented with 1% Iso Vitalex (BBL Microbiology Systems, Cockeysville, MD, USA), 5% fetal calf serum (Gibco, Paisley, Scotland), 1% haemoglobin (Difco, Detroit, MI, USA) and 3 [mu]g/ml vancomycin. The second medium had a gonococcal agar base (GC-Agar, Difco, Detroit, MI, USA) and the same supplements as the first. Isolates were identified on the basis of typical colony morphology and Gram-stain results. Specimens for isolation of herpes simplex virus (HSV) were obtained with a nontoxic cotton swab which was transported in Hank's balanced salt solution to the Institute of Tropical Medicine, Antwerp, Belgium, where HSV was detected by its cytopathic effect on a monolayer of Vero cells. …