Correlation of hepatobiliary ultrasonographic findings with cd 4 cell count and liver enzymes in adult hiv / aids patients in Jos

1 Department of Radiology, Faculty of Medical Sciences, Jos University Teaching Hospital, Jos, Nigeria. 2 Department of Radiology, Federal Medical Centre, Keffi, Nasarawa State, Nigeria. 3 Department of Physiology, Faculty of Biomedical Sciences, Kampala International University, Western Campus, Ishaka-Bushenyi, Uganda. 4 Department of Anatomy, Faculty of Biomedical Sciences, Kampala International University, Western Campus, IshakaBushenyi, Uganda.


INTRODUCTION
Human immune deficiency virus (HIV) is a retrovirus that stores its genetic material as ribonucleic acid (RNA).Two strains of HIV are known to infect humans; HIV-1 and HIV-2.HIV-1 is by far the most common pathogen worldwide (Sebastian, 2002;Obajimi et al., 2008).In 2013, it was estimated that the number of people living with HIV worldwide were more than 35 million; 70% of these individuals were in sub-Saharan Africa were in sub-Saharan Africa and the majority (61%) were women.Southern Africa is the worst affected region with the adult HIV prevalence exceeding 17.3% in eight southern African countries in 2011 (Igbinedion et al., 2009;Global Fact sheet, 2012;Worldwide AIDS and HIV statistics, 2011).The individuals who require antiretroviral therapy (ART) may not have access to it in sub-Saharan Africa thus worsening the morbidity and mortality.The prevalence of HIV infection in Plateau state is 4.9% while North Central Nigeria has the highest prevalence of 6.1%, the National prevalence is 4.4% (World Bank Nigeria Report, 2008;Federal Ministry of Health, 2005).Liver enzymes [alanine/aspartate transaminases (ALT/AST), and alkaline phosphatase (ALP)] abnormalities have been shown to be common in HIV/AIDS patients, probably due to direct inflammation induced by HIV virus on the liver cells and gall bladder or by other opportunistic agents (Ejilemele et al., 2007).Hepatobiliary disease is an increasingly important cause of morbidity and mortality in patients with HIV/AIDS (Gore et al., 1998).With the availability of highly active antiretroviral therapy (HAART), there has been an observable change in the pattern of liver disease in these patients (Goldin et al., 2002).Although opportunistic infections and neoplasm are still seen, co-infection with hepatitis viruses, especially HCV, is now emerging as the most significant cause of liver disease in this group of patients (Guardaat el., 1983).In addition, drug-induced liver damage is becoming more prevalent due to the increased complexity and toxicity of the HAART regimens used (Goldin et al., 2002).Biliary disorders seen in AIDS patients can be classified into AIDS cholangiopathy (AC), acalculous cholecystitis (ACC) and non-HIV associated disease such as gallstones (Flum at el., 1997).Gallstone disease is the most commonly observed cause of acute cholecystitis in this population (Flum at el., 1997).While HIV related biliary disorders are rare, they are associated with significant morbidity when they occur.AC and ACC were first reported in 1983 by Guarda and Pitlik respectively (Guarda et al., 1983;Pitlik et al., 1983).The two diseases can occur concurrently with opportunistic infections.Bile ducts in AIDS patients appear to be uniquely susceptible to opportunistic infections (Enns et al., 2003).Since the advent of highly active antiretroviral therapy (HAART), the incidence of AIDS associated biliary disorders have been steeply declining (O" Hara et al., 2009;Ko et al., 2003).Distinct hepatobiliary sonographic findings have been reported in these patients and include hepatomegaly, altered liver echotexture, gall bladder sludge, gall bladder wall thickening, gallstones, AIDS Cholangitis and periportal lymphadenopathy (Pawar et al., 2013).CD 4 cell count is one of the baseline measurements and an important parameter used in the assessment of AIDS progression and follow-up of AIDS patients.CD 4 cell depletion in HIVinfected patients results from ongoing viral replication.Hepatobiliary diseases were initially thought to be associated with advanced immunosuppression (CD 4 cell counts < 200 cells/microlitre), but it was later recognized that the lesions associated with the disease can occur at any stage of HIV-1 infection, even before antibody seroconversion (Herman et al., 2003).The relationship of hepatobiliary sonographic findings with CD 4 cell count and liver enzymes has not been well exploited in this environment.

Study design, period and setting
This was a hospital based cross-sectional study that spanned over the period of six months from November 2015 to April 2016 in the Department of Radiology, Jos University Teaching Hospital (JUTH), a tertiary health institution situated in the central part of Jos, Nigeria.
JUTH is one of the teaching hospitals in the North-Central Zone of Nigeria.The hospital provides medical and surgical care, gynaecology and obstetrics services along with radiology, emergency services and an outpatient clinic to the population of the city and neighboring states of Bauchi, Gombe, Benue, Kogi, Nasarawa, Taraba, Adamawa and parts of Kaduna State.The hospital has an outpatient daily clinic (Clinic II) for HIV/AIDS management.

Ethical approval and consent to participate
Ethical clearance was obtained from the Research and Ethical Committee of Jos University Teaching Hospital, Jos, Nigeria dated 27 th March 2015 with authorization: JUTH/DCS/ADM/127/XIX/6074. In addition, informed consent was obtained from each enrolled participant.

Target population/study participants
The study population comprised patients aged 18 years and above confirmed with HIV who were referred for abdominal ultrasound scan from Clinic II located within the Jos University Teaching Hospital complex.Clinic II is a specialized clinic exclusively for the management of HIV/AIDS patient.

Study inclusion criteria
Participants in this study were those patients aged 18 years and above confirmed with HIV infection.Participants were selected being informed of the nature of the study, the potential benefits and safety of the procedure, as well as the voluntary nature of the participation.

Study exclusion criteria
Participants who were pregnant HIV positive patients, HIV positive adults with confirmed co-infection with Hepatitis A, B or C viruses and HIV positive adults with co-existing medical conditions such as Diabetes mellitus, sickle cell disease and hypertension were excluded from the study.

Data collection procedure and analyses
Patients were prepared by asking them to fast for 6 to 8 h to reduce bowel gas and ensure gallbladder distension.Blood pressures were measured using a standard mercury sphygmomanometer to exclude hypertensive patients from the study.Western Blot and CD4 + results were obtained from patients" case files.All the patients were on Highly Active Antiretroviral Therapy (HAART).The time between CD4 + count and ultrasound findings were less or equal to six months.Five millilitres (5 ml) of blood sample was collected from each patient using a new syringe into a clean dry glass test tube and centrifuged to obtain the serum.This serum was analysed for liver enzymes (ALT, AST and ALP) by enzymatic method using spectrophotometer.The ultrasound scan examination was performed using a high resolution real time ultrasound scanner (ALOKA SSD-3500, Co. Ltd., Tokyo, Japan, 2007) equipped with 3.5 MHz curvilinear transducer and Doppler facility.Ultrasound scan is operator dependent; therefore, all scanning and measurements were done only by the researcher in order to minimize inter-observer error.All measurements were taken twice to reduce intra observer variability.The liver, gallbladder and biliary ducts were then examined sonographically using standard ultrasound scanning procedures.

Data preparation, management and analysis
At the end of the examinations, the hepatobiliary ultrasonographic findings and the data obtained were entered into a computer and processed using SPSS for window version 20.0 (Microsoft ® Inc.Chicago, Illinois, USA.2011).Mean ± standard deviation was used to summarize variables.The variables were tested for correlation via the Chi-square test, students" test and cross tabulations.P value of 0.05 or less was considered statistically significant.

Sex and age variables
A total of 410 HIV positive patients, who met the inclusion criteria were studied.This comprised 140 males (34.15%) and 270 females (65.85%) with male to female ratio of approximately 1:2.Majority of the male patients were within age range 41-50 years while most female patients presented at an earlier age range of 31-40 years.The mean age was significantly higher in males than females (Table 1).

CD4 count and liver enzymes analyses
The frequency of patients reduced with increasing severity of disease and decreasing CD 4 + count from 47.6% for "not significant" (≥500 cell/microlitre) to 9.3% for "severe" (<200 cell/microlitre) CD 4 + count (Table 2).The result showed 69.8% of patients had normal serum ALT; while 30.2% had various degrees of elevated ALT.The highest abnormal frequency for ALT was seen in Grade 1 (19.8%) and the lowest (1.5%) in grade 3 (Table 2).Most patients (51.7%) had normal serum AST while 48.3% had abnormal levels of serum AST.The highest abnormal frequency for AST was also seen in Grade 1 (22.2%) and the lowest (1.5%) in grade 3 out of this 22.2% was seen in grade 1 (Table 2).Majority of patients (60.5%) had various grades of abnormal ALP with the highest abnormality seen in Grade 0 (32.4%) and the lowest in Grade 2 (2.5%) (Table 2).No patient had Grade 3 ALP abnormality.

Age and CD4 count analysis
The CD 4 + count decreased with increasing age in all CD 4 classification groups, except for age range 31-40 years and >50years which showed no specific trend with CD 4 + count.There was significant negative correlation between age and CD 4 + count (P=0.001)(Table 3).

Sex and CD4 count analysis
For both genders, the frequency of patients reduced with increasing severity of disease and decreasing CD 4 + count.Most of the patients (47.6%) were in the "not significant" disease group (Table 4).100) 104( 100) 69( 100) 410( 100)

Age, sex and liver enzymes analysis
Across all the age groups, ALT was normal in 69.8% of patients.Of these, the ALT abnormality was mostly Grade 1 seen in 19.8% and least frequent abnormality (1.5%) was grade 3.There was no significant association between patients age and serum ALT (P=0.063)(Table 5).About half (51.7%) of patients had normal AST and the proportion of patients with normal values decreased with increasing age group, this finding was statistically significant (P=0.018,Table 5).The most frequent AST abnormality was grade 1 and least grade 3 seen in 1.5% of subjects.Only about 40% of subjects had normal ALP and the proportion of patients with normal values decreased with increasing age group, this finding was statistically significant (P=0.019,Table 5).The most frequent ALP abnormality was grade 0 seen in 32.4% and least was grade 2 seen in 2.4% of subjects.In both genders and for all the enzymes, the frequency of patients generally reduced with increasing grade of liver enzymes abnormalities.This finding was not statistically significant (P=0.995,0.97 and 0.970 respectively (Table 6).

Ultrasound findings
Two hundred and sixteen (52.7%) patients out of the 410  GB=Gallbladder.*Some patients presented with more than one ultrasound findings.
studied had normal hepatobiliary ultrasound scan while the remaining 194 patients had various sonographic abnormalities (Table 7).The commonest hepatobiliary abnormality seen was enlarged gall bladder followed by increased liver Echogenicity and hepatomegaly (Figure 1) while the least frequent abnormal finding was gall stones (Figure 2) seen only in female patients.No significant association was seen between sonographic abnormalities and gender (P = 0.507) (Table 8).

Age and abnormal ultrasound findings
Hepatomegaly and increased liver echogenicity were the commonest abnormalities seen in majority of patients over >50 years while thickened gallbladder wall and gallbladder sludge were more frequent in age group 41-50 years.Enlarged gallbladder was seen commonly in younger age group of ≤30 years.There was statistically significant association between the abnormal ultrasound     findings and age (P=0.005)(Table 9).

CD4 count and Ultrasound findings
Of the patients studied, 52.7% had normal ultrasound findings.In those with abnormal ultrasound findings, 43.8% were in the "not significant" CD 4 group and the frequency of abnormal ultrasound findingsreduced with decreasing CD 4 + count and increasing disease severity, implying a negative correlation with disease severity (P = 0.001) (Table 10).
In those with abnormal ultrasound findings, patients with hepatomegaly and enlarged gallbladder were commoner in the mild CD 4 + class while those with increased liver echogenicity, thickened gallbladder wall and sludge were most frequent in the severe CD 4 + category.Gall stone was present only in the not significant CD 4 + class.Enlarged GB, increased liver echogenicity and hepatomegaly correlate significantly with CD 4 + count (P = 0.001).However, thickened gallbladder wall and gallbladder sludge did not individually correlate significantly with CD 4 + count (P = 0.12 and 0.57 respectively) (Table 11).

Liver enzymes and ultrasound findings
Majority (46.1%) of the patients who had normal ultrasound finding had normal serum liver enzymes however, some patients with normal serum liver enzymes also had abnormal ultrasound findings and vice versa.Hepatomegaly was seen in various grades of liver enzymes abnormalities except grade 2 ALT.It was most frequent in grades 3 for ALT and AST but in grade 1 for ALP.This finding was statistically significant (P = 0.001) (Table 12).Increased liver echogenicity varied with degree of elevated liver enzymes but was worse in grade 2 for ALT, grade 3 for AST and grade 2 for ALP.This finding was also statistically significant (P = 0.001, Table 12).Enlarged GB was seen in all grades of elevated liver enzymes and was most frequent in grade 2 for ALT, grade 3 for AST and grade 1 for ALP.This finding was statistically significant (P = 0.001, Table 12).Thickened GB wall was seen in grades 0 and 1 for ALT, and grades 0-2 for AST and ALP.The finding was statistically significant (P = 0.013 and 0.017 for ALT and AST respectively) (Table 12).
Among the three (3) liver enzymes studied, ALP has the highest sensitivity (68) to truly detect those patients with sonographic detectable hepatobiliary disease followed by AST (55.7).However, the ALT has the highest specificity (75.9) to detect correctly absence of hepatobiliary diseaseon ultrasound.ALT also has the highest positive predictive value implying those with abnormal ultrasound truly have the hepatobiliary disease while ALP has the highest negative predictive value implying that those with normal ultrasound truly have no hepatobiliary disease (Table 13).

Logistic analysis
Logistic regression shows that hepatomegaly was the most likely abnormal ultrasound finding (OR=4.08) to be seen in HIV/AIDS patients with advanced/severe CD 4 counts followed by enlarged gallbladder (OR=2.83)while increased liver echogenicity was the least probable.However these findings were not statistically significant with P = 0.08 or 0.48 (Table 14).Among the three (3) liver enzymes studied, abnormal ALP enzyme value (OR=1.3)was the most likely to be seen in patients with advanced/severe CD 4 counts followed by AST (OR=0.63).However these findings were also not statistically significant with P = 0.61 or 0.34) (Table 15).

DISCUSSION
The number of people living with HIV/AIDS worldwide was estimated to be 36.9 million at the end of 2014 according to World Health Organization (Worldwide and HIV statistics, 2011).The use of highly active antiretroviral therapy (HAART) has modified the disease Angbalaga et al. 91 pattern.Clinical and ultrasound findings in HIV/AIDS depend on the immune status of the patient as represented by their CD 4 + count (Igbinedion et al., 2009).Abnormalities of liver enzymes are common in HIV/AIDS patients and these may be due to direct inflammation induced by HIV on the liver cells and gall bladder or due to complications of antiretroviral drugs (Ejilemele et al., 2007).A total of 410 HIV positive patients on HAART were studied, majority of the patients were females (65.8%) with male to female ratio of 1: 2. This is in agreement with a study done in South Western Nigeria by Obajimi et al. (2008) which documented a female preponderance of 66.5%.This finding is also similar to the report of the Joint United Nations Programme on HIV/AIDS (UNAID), 2010) which documented that the percentage of women living with HIV/AIDS increased from 43% in 1999 to 50% in 2010 and that Sub-Sahara African women comprised 59% of adults infected with HIV (Global Fact sheet, 2012).The sex ratio in this study was however at variance with the study documented in Sudan by Mahmoud et al. (2012) which showed a higher male preponderance of 66%.The finding of female preponderance in this study could be due to the fact that females" genital anatomy places them more at risk for sexually transmitted infections, compared to men (Center for Disease Control and Prevention Fact sheet, 2017).Females also have better health seeking behaviours and are routinely counseled during antenatal care for HIV testing leading to the discovery of their HIV status (Center for Disease Control and Prevention Fact sheet, 2017).
Approximately 83% of the patients were within the age range of 18 to 47years, thereby constituting the majority of the patients in this study population.This age range was described by Adeoye (2005) as the economically productive segment of Nigerian society and also the age group at the greatest risk of HIV/AIDS.However, there were few patients above 50 years old, possibly because most of the older populations are less sexually active (Adeoye, 2005).Finding is in agreement with a study done by Igbinedion et al. (2009) in Benin-Nigeria which documented about 89% of the patients to be between 18 to 47 years.The mean age for females and males were 37.59 years and 42.84 years respectively.This mean age was also consistent with the study done in South Western Nigeria by Obajimi et al. (2008) which documented mean age of female HIV patients to be 38.02years.This may be due to the fact that females are sexually active at an early age compared to males (Abubakar et al., 2014).
Based on the World Health Organization (WHO, 2006) classification of CD 4 + immunological profile in adult HIV infected patients, 73.4% of the patients had CD 4 + count of350 cells/µl and above making up 47.6 and 25.8% for the "Not significant" and "Mild" categories respectively while only 9.3% were in the "Severe" category ( that is CD4+ count<200cells/µl).This was at variance with the findings of Igbinedion et al (2009) who found almost half (46.3%) of the patients to be in the "Severe" CD4+ class.The higher CD4+ count and thus milder/not significant disease in this study maybe due to incessant public enlightenment, with resultant decreased stigmatization, earlier diagnosis and follow up as well as improved availability of anti retroviral drugs unlike in the past when most patients were well stigmatized, presented late and were unable to access or procure anti retroviral medications.This might also indicate the improved Based on the grading system developed by the Acquired Immune Deficiency Syndrome (AIDS) Clinical Trials Group (CTG) in assessing the serum level of liver enzymes of infected individuals.In this study, 53.7% of the patients had normal liver function tests while 46.3% of the patients had various abnormalities of their liver enzymes (ALT, AST and ALP).This is not in agreement with the study done by Ejilemele et al. (2007) in Port Harcourt Nigeria who documented a higher (87.6%) abnormal liver enzyme.This study is also at variance with the finding of Savita et al. (2015) in India which also documented a higher (63%) abnormal liver function tests (LFT).In this study, out of the patients with abnormal (46.3%) liver enzymes, 39.4% were classified as hepatocellular liver (elevated AST and ALT) injury while 60.6% were classified as cholestatic liver (elevated ALP) injury.This is not in agreement with the study done by Ejilemele et al. (2007) in Port Harcourt Nigeria who recorded a lower (14.5%)cholestatic liver injury and a higher (85.5%) hepatocellular injury (Ejilemele et al (2007).This study is also not in agreement with the finding of Savita et al (2015) in India which documented 63% of their patients with abnormal liver function tests (LFT).They recorded a lower (6%) cholestatic liver injury, a higher (29%) hepatocellular injury and 28% had a mixed pattern of liver injury.The reason for the higher prevalence of abnormal liver function tests in the Port Harcourt and India studies is most likely due to the fact that the patients studied were both not on antiretroviral therapy at the time of data collection.
Different hepatobiliary sonographic abnormalities such as enlarged gallbladder (GB), increased liver echogenicity, hepatomegaly, thickened GB wall, gallbladder sludge and stone were observed in this study which were comparable with those documented by other authors (Obajimi et al (2008); Igbinedion et al (2009); Pawar et al. (2013); Mahmoud et al (2012).Some of these abnormal ultrasound findings (enlarged GB, increased liver echogenicity and hepatomegaly) correlated significantly with CD 4 + count (P = 0.001).The "Not significant" CD 4 + class recorded 43.8% abnormal hepatobiliary ultrasound findings as compared to 14.4% noted in the "severe" CD 4 + class.In a similar study, Pawar et al. (2013) documented 43.3% abnormal hepatobiliary ultrasound findings in the "Not significant" CD 4 + (>350 cells/µl) category and 56.7% in the "Significant" CD 4 + (≤350 cells/µl) class.This was in contrast with the finding of Igbinedion et al. (2009) who documented a higher abnormal abdominal ultrasound finding of 85.7% in the "Not significant" CD 4 + class and 89.9% in the "severe" CD 4 + category.Hepatomegaly was the third most common abnormal hepatobiliary sonographic findings in this study as seen in 23.7% patients.Different studies had documented figures such as 13.3, 39.0, 40.0, 35.0 and 41.0% respectively for hepatomegaly; (Obajimi et al., 2008;Igbinedion et al., 2009;Javier et al., 2005;Tshibwabwa et al., 2000;Grumbach et al., 1989).According toIgbinedion et al. (2009), hepatomegaly in these patients could be due to infections, non specific response to infective hepatitis, fatty infiltration or neoplastic infiltration from lymphoma or Kaposi sarcoma.Increased liver echogenicity was the second most common finding as seen in 26.3% patients.Grumbach et al. (1989) documented a higher finding of 45.5%.Both hepatomegaly and increased liver echogenicity negatively correlate with CD 4 + count (Both P = 0.001).This was also in contrast with the study of Igbinedion et al. (2009), whose finding (hepatomegaly and increased liver echogenicity) did not correlate with CD 4 + count.Increased liver echogenicity in HIV could be due to fatty infiltration, chronic hepatitis, drug and alcoholic induced liver diseases as in the general population (Valentina and Giuseppe, 2012).Gallbladder enlargement was noted in 33.3% of patients while thickened gallbladder wall, gallbladder sludge and gallbladder stones were noted in 7.8, 6.6 and 2.1% of the patients respectively.In similar studies, Igbinedion et al. (2009) documented thickened gallbladder wall in fewer patients (7.3%) and gall stones in a higher proportion (4.7%) of the patients.Pawar et al. (2013) who recorded thickened gallbladder wall and gall stones in a higher proportion(9.0%)and (4.0%) of the patients respectively and gallbladder sludge in fewer patients(6.0%)which did not correlate with CD 4 + count.Obajimi et al. ( 2008) recorded gall stones in only 1.3% of their patients.Grumbach et al. (1989) recorded thickened gallbladder wall in 55%, dilated gallbladder in 18%, biliary sludge in 23% and gall stones in 5% of the patients.This difference (lower values in this study) in gallbladder abnormal findings may be attributed to the fact that there is better awareness on HIV/AIDS.Therefore, most patients present early coupled with the availability of HAART and early treatment.Thickened gallbladder wall in HIV could be due to cholecystitis, hepatitis, sepsis and tuberculosis (HIV infection and Gallstones-Risk factors, 2017).The major risk factors for developing gallstones in HIV include male sex, increasing age (>40years), those on Norvir therapy and high blood pressure (Romano et al., 1988).Gallbladder sludge in HIV is associated with opportunistic infection, cholecystitis and hepatitis (HIV infection and Gallstones-Risk factors, 2017).Hepatomegaly was the most likely abnormal ultrasound finding in HIV/AIDS patients with advanced/severe CD 4 counts while increased liver echogenicity was the least probable.However these findings were not statistically significant (P=0.77and 0.48 respectively) using logistic regression test.No specific reason could be adduced for this finding and further studies would be required in this regard.Aspartate and Alanine transaminases are excellent markers of hepatobiliary injury while Alkaline Phosphatase is a marker of biliary diseases; however, none of the enzymes are specific for hepatobiliary diseases as similar elevation of these enzymes may be encountered in myocardial infarction and some muscle injuries (Ocama et al., 2008).
In this study, the prevalence of abnormal liver enzymes (ALT, AST and ALP) and ultrasound abnormalities (hepatomegaly, increased liver echogenicity, enlarged GB, thickened GB wall, GB sludge and stone) in adult HIV positive patients was 47.3%.This was lower than in earlier reported studies done by Igbinedion et al. (2009), Pawar et al. (2013) and Mahmoud et al. (2012) which recorded 66.6, 96.0 and 93.3% respectively.This low prevalence may be due to the fact that all the patients at the time of conducting this study were already on regular Highly Active Antiretroviral Therapy (HAART).Some patients with normal liver enzymes had abnormal ultrasound findings and vice versa; an average of 46.3% of the patients had various grades of liver enzymes abnormalities.This was not in agreement with a study done by Savita et al 2015 in India which documented a higher (63%) abnormal liver function tests (LFT).In this study, 47.3% of the patients had various abnormal hepatobiliary ultrasound findings and this was in agreement with the study of Savita et al. (2015) who recorded 47% of various abnormal ultrasound findings in their patients.In this study, increased liver echogenicity was seen in 27.0% which is inconsistent with the finding of Savita et al. (2015) who recorded a higher percentage of 46.3%.This study recorded hepatomegaly in 27.1% of the patients which is also in contrast with a study done by Savita et al. (2015) who recorded 15.5%.In this study, both increased liver echogenicity and hepatomegaly significantly correlate individually with liver enzyme.There was however no correlation between LFTs (ALT, AST and ALP) and ultrasound findings in the study done by Savita et al. (2015).Infection by HIV increases activities of the three enzymes, which may be due to liver cells apoptosis caused by HIV infection, or immune response to HIV replication which subsequently leads to hepatocellular necrosis and inflammation (Housset et al., 1990;Hufert et al., 1993;Gendrault et al., 1991).Low frequency of abnormal liver function tests and various abnormal hepatobiliary ultrasonographic findings in this study could be due to early diagnosis and treatment with readily available antiretroviral therapy.Elevated serum alkaline phosphatase was the most likely liver enzyme abnormality in patients with advanced/severe CD 4 counts followed by aspartate transaminases as indicated by logistic regression analysis but this was not statistically significant (P = 0.61).

Conclusion
Ultrasonography is a versatile tool for evaluating hepatobiliary organs affected by HIV/AIDS.Most abnormal sonographic hepatobiliary findings in patients with HIV/AIDS correlated significantly with CD 4 + count and liver enzyme levels.Therefore, ultrasound scan could be introduced as part of management/follow-up of HIV/AIDS patients as it may improve management out come in these patients.Ultrasound is easy to perform, non invasive, inexpensive and readily available in this part of the world where HIV/AIDS is most prevalent.Among the liver enzymes studied, ALP has the highest sensitivity to truly detect those patients with sonographic Detectable hepatobiliary disease and highest negative predictive values implying those with negative abnormal ultrasound truly have no hepatobiliary disease.

RECOMMENDATIONS
Hepatobiliary ultrasound scan should be made an initial routine examination in all HIV clinics for better management of these patients.

Figure 1 .
Figure 1.Longitudinal ultrasound image of the liver demonstrating increased liver echogenicity and hepatomegaly in a 43 year old female HIV positive patient.

Figure 2 .
Figure 2. Longitudinal ultrasound image of the gallbladder (GB) showing a gallstone (White arrow) in a 39 year old HIV female positive.

Table 1 .
Demographic characteristics of HIV/AIDS patients in Jos.

Table 2 .
Distribution of patients in relation to CD4 + count and liver enzymes.

Table 3 .
Relationship between age and CD4 + count of HIV/AIDS patients frequency (%).

Table 5 .
Relationship between age and Liver enzymes of HIV/AIDS patients.

Table 6 .
Relationship between gender and Liver enzymes of HIV/AIDS Patients.

Table 7 .
Distribution of abnormal hepatobiliary ultrasound findings in HIV/AIDS patients.

Table 9 .
Distribution of abnormal Ultrasound findings in relation to age group frequency (%).

Table 10 .
Relationship between ultrasound findings and CD4 + classification.

Table 11 .
Distribution of Ultrasound findings in relation to CD4 + classification frequency (%).
*Some patients have multiple findings.GB = Gallbladder.

Table 12 .
Distribution of Ultrasound findings in relation to liver enzymes classification.

Table 13 .
Sensitivity, Specificity, Positive predictive value and Negative predictive value using ultrasound as outcome variables.

Table 14 .
Logistic regression abnormal ultrasound findings and advanced/severe CD4 counts.

Table 15 .
Logistic regression liver enzymes and advanced/severe CD4 counts