The Effect of Low CD4+ Lymphocyte Count on the Radiographic Patterns of HIV Patients with Pulmonary Tuberculosis among Nigerians

Objective. To assess the radiographic features in patients with Human Immunodeficiency Virus (HIV) complicated by pulmonary tuberculosis (PTB), and the association with CD4 lymphocyte count and sputum smear. Method. A prospective study was carried out on 89 HIV positive patients with PTB. The demographics, smoking history, sputum smear result, chest radiographic findings and CD4 lymphocyte count were documented. Results. Out of the 89 patients recruited in the study, 41 were males and 48 were females. Eighteen (18) patients had typical radiographic features, 60 patients had atypical radiographic features while only 11 of them had normal radiographic films. Sixty eight (68) patients had CD4 count <200 cells/mm3, 19 patients had CD4 count between 200–499 cells/mm3, while only 2 patients had CD4 count from 500 cells/mm3 upwards. The association between low CD4 count and radiographic finding was statistically significant, (P value <0.05). Sixty (60) patients had negative sputum smear for Acid and Alcohol Fast Bacilli (AAFB), while the remaining 29 patients had positive smear. The association between low CD4 count and negative smear was statistically significant (P value <0.05). Conclusion. The radiographic pattern and the result of the sputum smear for AAFB has a significant relationship and association with the immune status of patients with Human Immunodeficiency Virus (HIV) complicated by pulmonary tuberculosis.


Introduction
Human immunode�ciency virus (HIV) is a potent risk factor for tuberculosis (TB), both through an increase in the reactivation of the latent Mycobacterium tuberculosis infection and through an accelerated progression from infection to active disease, by undermining the cell-mediated immunity through depletion of CD4 lymphocytes [1][2][3][4].
TB has a great impact on morbidity and mortality in HIV-1 infected individuals than all other opportunistic infections [3]. TB and HIV infections have a synergistic in�uence on the host immunoregulation. TB can develop at any stage of immunosuppression regardless of the level of the circulating CD4+ T-lymphocytes [4]. CD4+ lymphocytes count is one of the surrogate markers for evaluating the degree of immunosuppression and HIV disease progression [4].
e levels of circulating CD4+ lymphocytes has a great impact on the radiographic pattern of TB. In HIV infections, TB can produce both typical and atypical radiographic patterns depending on the degree of immunosuppression [5][6][7][8]. Atypical radiographic presentations are lower frequency of cavitations, higher frequency of mediastinal lymphadenopathy, lower lung �one in�ltrates, and even a normal chest radiograph. Typical presentations include upper lobe �brosis, 2 Tuberculosis Research and Treatment bilateral in�ltrates, consolidation, and cavitations [6][7][8]. us patients with low CD4 lymphocyte count have more of features of primary TB, while those with a high CD4 count will have features of postprimary TB.
is study was undertaken to determine the effect of low CD4+ lymphocyte count on the radiographic patterns of HIV patients with pulmonary TB among Nigerians. ere is dearth of studies on this subject in Nigeria.

Materials
2.1. Study Setting. e study was done at University of Benin Teaching Hospital from April to May 2007. is hospital is a government designated centre of excellence for the management of HIV/AIDS. e patients for the study were recruited from the HIV clinic of this hospital. ey were con�rmed HIV positive patients with pulmonary TB.

Inclusion Criteria
(1) HIV positive patients with pulmonary TB aged 18 years and above who presented at the clinic within the study period and who had not commenced drug treatment.
(2) All HIV positive patients with pulmonary TB who consented to join in the study.
(2) Patients with other immunosuppressive diseases like diabetes mellitus, malignancies, and so forth.

Method of Data Collection
is was a prospective study done from April to May 2007. Informed consent was obtained from the patients. Information was gathered from the patient's case note. e subject's biodata and the results of the following investigations were collected.
e chest radiographs were blindly reported by two radiologists who were not aware of the enrolment into the study. e two reached a consensus on radiological �ndings of a radiograph when there is a disagreement in their evaluation. ey evaluated the radiographs for mediastinal lymphadenopathy, in�ltrates, cavitations, pleural effusions, and localized or military shadows and also determined the predominantly affected lung zones. e chest radiograph results were grouped into three namely, those with typical features, those with atypical features, and those with normal �lms. e sputum smear result was grouped into two namely, those positive for AAFB and those negative for AAFB. e CD4 lymphocyte cell count results was also grouped into three namely, CD4 count <200 cells/mm 3 , CD4 count 200-499 cells/mm 3 , and CD4 count from 500 cells/mm 3 upwards.

Method of Data Analysis
e statistical package for social sciences (SPSS) version 16.0 was used to record and analyse the data. Frequency tables were drawn to show the distribution of data within variables. Contingency tables were drawn to compare two discrete variables. Pearson Chi-square was used to test signi�cance. A value < 0.05 was considered signi�cant.
�. Case De�nition e diagnosis of PTB/HIV coinfection was based on criteria for diagnosing TB in poor resource settings where there are no facilities and manpower for mycobacterium tuberculosis culture: (a) the diagnostic criteria of TB given in the World Health Organisation (WHO) treatment of tuberculosis guideline for national programmes [5]� (b) speci�city of clinical criteria in diagnosing TB patient [6,9].
More of the patients (74 patients (83.15%)) had never smoked, while a small number (15 patients (16.85%)) had smoked at one time or the other in their lives. Sixty (60) patients (67.42%) had negative sputum smear for acid and alcohol fast bacilli, while 29 patients (32.58%) had positive smear.
Out of the 60 patients that had negative sputum smears, only 2 of them had CD4 count ≥200 cells/mm 3 . e other 58 patients with negative smears had CD4 count <200 cells/mm 3 , and the association between severe immunosuppression (CD4 count <200) and negative sputum smear was signi�cant ( value < 0.05) ( Table 3).

Discussion
Results from the study showed a signi�cant relationship between the CD4 lymphocyte cell count and the radiographic features of HIV positive patients with pulmonary tuberculosis (Table 2). e CD4 count is an indicator of immune status and stage of HIV infection. Severe immunosuppression and CD4 count <200 cells/mm 3 were signi�cantly associated with the presence of mediastinal lymphadenopathy. is is in keeping with other studies worldwide [9][10][11]. Other features of primary TB (atypical features) like middle and lower lung zone involvement, military pattern, and normal �lms were also more common in patients with CD4 count <200 cells/mm 3 , (Table 2). Cavity formation and other features of postprimary TB on chest radiograph were found to be common and signi�cantly associated with more immunocompetency (CD4 count ≥200 cells/mm 3 ). Several other studies on the association between cavity formation and high CD4 count con�rm this [11,12]. Formation of cavities in TB infection requires an adequate delayed type of hypersensitivity reaction and an intact cell-mediated immunity in the host. Typical features (upper lung zone involvement) are seen more oen in HIV patients with less immunosuppression than in those with severe immunosuppression. It is not surprising that HIV patients with high CD4 counts will have upper lung zone involvement because TB in the upper lung zone is usually common in HIV negative patients with high CD4 count and thus immunocompetent.
Only 11 out of the 89 patients in the study had normal chest radiographs, and all the 11 patients had CD4 count <200 cells/mm 3 . e association between severe immunosuppression and normal radiograph was signi�cant ( Table  2). is was supported by another study [13] but was not supported by a study done in �SA which found no signi�cant association between CD4 count and normal chest radiograph [11].
e �nding of a normal chest radiograph and negative sputum smear microscopy, in HIV patients coinfected with pulmonary TB, poses a great challenge for the diagnosis of pulmonary TB in poor resource countries, where facilities for culture of mycobacterium tuberculosis are scanty or nonexistent, and thus diagnosis requires high index of suspicion. More studies are required to con�rm if the absence of radiographic �ndings represents early stages of either primary TB or a reactivation, or one caused by intrathoracic lymphadenopathy unable to be detected by simple radiographic examination [11,12]. In a cross tabulation between sputum smear and chest X-ray �ndings, it was found that T 4: Relationship between sputum smear and chest X-ray features.
most of the patients that had atypical chest X-ray features also had negative sputum smears (Table 4). e relationship between these two was signi�cant. is �nding con�rms that in severe immunosuppression, the usual immune response to TB infection is no longer maintained. In this study, atypical or primary pattern was common in patients with CD4 count <200 cells/mm 3 , while typical or postprimary pattern was commoner in those with CD4 count >200 cells/mm 3 and commonest in those with CD4 count ≥500 cells/mm 3 . is was similar to what was found in some other studies [10,11]. e different radiographic appearances have different pathogenesis, and these were greatly modi�ed by the level of CD4 count (degree of immunosuppression) and cell-mediated immunity.
is study has revealed that various radiographic manifestations of HIV coinfected with pulmonary TB are related to the level of immunosuppression (CD4 count). Physicians need to be aware of this �nding, and those in countries with poor resources should have high index of suspicion to be able to make proper diagnosis.