Aetiology and presentation of pulmonary disease in HIV-Infected patients at risk of early mortality at the University Teaching Hospital in Lusaka, Zambia

Abstract

Introduction: Pulmonary disease is a leading cause of morbidity and mortality in HIV-infected patients. Yet specific causes are often unknown in third-world settings due to a lack of extensive diagnostic facilities, particularly in patients with sputum smear negative for acid-alcohol fast bacilli (AAFB). Fibre-optic bronchoscopy (FOB) is available as a diagnostic tool at the University Teaching Hospital. It has been established as an accurate, reliable and safe diagnostic means for pneumonias in severely immunosuppressed patients. Objectives: To determine the aetiology and presentation of pulmonary disease in Highly Active Anti-Retroviral Treatment (HAART)-naïve, HIV-infected Zambian adults with severe immunosuppression Design: A cross-section study Methods: Our study algorithm comprised initial sputum screening with Ziehl-Neelsen stain for HAART-naïve HIV-infected patients with CD4+ counts less than 200/μL presenting with pulmonary symptoms. Those who were unable to expectorate sputum and those in whom sputum smears were negative for AAFB, underwent bronchoscopy. Bronchoalveolar lavage (BAL) specimens were collected and assessed for bacteria, fungi, Mycobacteria and Pneumocystis jirovecii. Microbiological diagnoses were correlated with clinical and radiological findings. Results: Of 113 enrolled patients, 43 (38.1%) had sputum smears positive for AAFB. 53 (46.9%) had smears negative for AAFB and 17 (15.0%) were unable to expectorate sputum. 58 of the 70 (82.9%) sputum AAFB-negative or sputum-scarce patients agreed to further screening with bronchoscopy. Seven (12.1%) of the BAL specimens were positive for TB on smear, while 14 (24.1%) had TB diagnosed on culture alone. Cumulatively, 64 (56.6%) patients were diagnosed with TB. Two (1.8%) patients had Mycobacteria intracellulare and one (0.9%) had Mycobacterium avium complex cultured on BAL. Pneumocystis jirovecii was found in five (4.4%) patients, Candida species in six (5.3%), Klebsiella in five (4.4%) and gram negative enteric bacteria in two (1.8%). One (0.9%) each of Staphylococcus aureus, Streptococcus pneumonia and Proteus mirabilis were cultured. Additionally, Kaposi’s sarcoma was diagnosed in three (2.7%) patients on visual inspection at bronchoscopy. One (0.9%) patient had both TB and PJP. The cause was undetermined in 34 (30.1%) patients. Miliary infiltrates on chest radiograph was associated with TB (χ² 5.353; p=0.02). TB was further associated with micro-nodular (χ² 4.557; p=0.03) and nodular (χ² 7.864; p=0.01) infiltrates, as well as bilateral hilar lymphadenopathy (χ² 4.105; p=0.03) and haemoglobin less than 8g/dL (χ² 6.160; p=0.01). Respiratory rate of 40/minute or more was associated with PJP (χ² 5.595; p=0.02). BAL smears for TB in sputum smear negative patients had a sensitivity of 33% and specificity of 100%. Conclusion: Mycobacterium tuberculosis is the commonest cause of pulmonary disease in our study population. Clinical and radiological correlates of TB can be used in the diagnosis of Aetiology and presentation of pulmonary disease in HIV-infected patients at UTH, Lusaka Kondwelani John Mateyo AAFB smear negative pneumonia, in the absence of bronchoscopy, which has proven to be a useful tool for diagnosis of AAFB smear-negative pneumonia.