Liver involvement in HIV-infected patients with early syphilis

Abstract

The aim of the paper is to analyse the prevalence of liver involvement and related factors in HIV-infected patients with early syphilis (<2 years). Liver involvement was defined as an elevation above normal ranges of alanine transaminase, aspartate aminotransferase, gamma-glutamyltransferase and/or alkaline phosphatase during early syphilis, or doubling of previous levels in patients with liver enzyme elevation before syphilis. We undertook a multicentre study and of the 147 cases, 86.4% were men who had sex with men, and the diagnoses of syphilis and HIV infection were coincident in 48 (32.7%). Liver involvement was detected in 45 (30.6%) and the only related factor was a rapid plasma reagin (RPR) titre ≥1/64 (odds ratio 3.76; 95% confidence interval 1.3–10.5; P = 0.012). In conclusion, liver involvement occurs in around one-third of HIV-infected patients with early syphilis and is associated with high RPR levels. Syphilis should be included in the differential diagnosis of liver enzyme elevation in HIV-infected patients.

Keywords

syphilis HIV liver enzymes hepatitis early syphilis

INTRODUCTION

Liver enzyme elevation is common in HIV-infected patients.^1,2 Although this elevation was traditionally due to opportunistic events like Mycobacterium tuberculosis, Cryptosporidium complex, lymphomas or Kaposi sarcoma, liver enzyme elevation since the introduction of combination antiretroviral therapy (cART) is more usually secondary to viral hepatitis, potentially hepatotoxic drugs (as are many of the antiretrovirals), alcohol and hepatic steatosis.^1–3 Liver involvement has also been seen in patients who are co-infected with syphilis during the early stages of the disease,^4–8 though only 23 cases of syphilitic hepatitis have been reported.^4,6,8–11 In fact, the incidence of acute hepatitis during early syphilis infection and the factors associated with syphilitic hepatitis among HIV-infected patients are currently unknown. Accordingly, we aimed to analyse the prevalence of liver involvement and related factors in HIV-infected patients diagnosed with early syphilis.

PATIENTS AND METHODS

We conducted a multicentre case series study of patients co-infected with HIV and syphilis. Databases from three Spanish hospitals were first reviewed to identify all HIV-infected patients who were diagnosed with early syphilis infection (duration <2 years) from January 2004 to December 2010. The diagnosis of syphilis was based on positive serological test results, including a non-treponemal antibody test and a specific treponemal antibody test. In addition, to be included in the study, all patients had to have had serological liver tests (transaminases, alkaline phosphatase and bilirubin) carried out around the time of the syphilis diagnosis and at least once more before and/or after the diagnosis of syphilis at the time points defined as follows: ‘before’ (3–9 months before the diagnosis of syphilis), ‘during’ (between 12 weeks before and 2 weeks after the diagnosis of syphilis) and ‘after’ (3–9 months after the diagnosis and treatment of syphilis). Patients with acute hepatitis due to viral hepatitis or any other cause during the episode of syphilis, patients without completed treatment for syphilis and those with missing data were excluded from the analysis. The changes in liver test results for the ‘before-to-during’ and ‘during-to-after’ periods for each patient who had paired data were calculated. Liver involvement was defined by rises above the normal range in the levels of aspartate aminotransferase, alanine transaminase (ALT), gamma-glutamyltransferase (GGT) and/or alkaline phosphatase during the syphilis infection or doubling of the previous value in patients with values that were already high before the diagnosis of syphilis. Various different episodes of syphilis in the same patient were included, provided correct treatment and response were confirmed for the first episode (with fall in rapid plasma reagin RPR titres to negative or less than or equal to half the RPR at treatment and epidemiological exposure to syphilis).

The qualitative variables are expressed in percentages and the associations were analysed with the Chi square test (χ²) or Fischer's exact test. The quantitative variables are expressed as the mean (interquartile range) and the differences were analysed with the Student's t-test after determining that the quantitative variables followed a normal distribution (Kolmogorov–Smirnov test). When there were more than two groups to compare, a one-way or multi-way analysis of variance (ANOVA) was used. The multivariate analysis for variables related with liver involvement was done with the Cox proportional hazards model. The odds ratios (ORs) were determined and 95% confidence intervals (CIs) calculated for significant variables. In all cases two-tailed tests were conducted and significance was set at P < 0.05. The analyses were done with SPSS 17.0.0 (SPSS, Inc., Chicago, IL, USA).

RESULTS

During the study period 155 episodes of syphilis were diagnosed; five cases were excluded due to lack of data and three due to coincident acute viral hepatitis. Therefore, the final study included 147 cases. Most (86.4%) were men who had sex with men (MSM), with a mean age of 38.7 years, and the diagnoses of syphilis and HIV infection were coincident in 48 (32.7%) patients. At syphilis diagnosis, the mean CD4 cell count was 497 cells/mm³, 70.7% of those with prior known HIV infection were on cART and 64.2% of these had an undetectable HIV viral load. Liver involvement during syphilis was detected in 45 (30.6%) patients, and only one of these presented with symptomatic hepatitis, mainly vomiting and jaundice. Liver involvement was more frequent in patients with a coincident diagnosis of HIV and syphilis, and in cases with any symptoms of syphilis (Table 1), but the only factor associated with this abnormality in multivariate analysis was an RPR titre ≥1/64 (OR 3.76; 95% CI 1.3–10.5; P = 0.012). No association was found with CD4 cell count, HIV viral load, chronic co-infection with hepatitis viruses or clinical presentation of syphilis (asymptomatic [early latent], chancre or secondary syphilis).

Table 1

Comparison of patients with and without liver involvement (univariate analysis)

	LI n = 45 (80.6%)	No LI n = 102 (69.4)	P
Age (years)	39.1 (31.8–46.6)	38.5 (30.4–45.2)	0.7
Co-infection HBV/HCV	10 (22.2)	16 (15.6)	0.2
On prior ART*	18 (75.0)	52 (69.3)	0.2
VL <50 copies/mL^†	9 (50.0)	36 (69.2)	0.4
CD4 cells/mm³	494 (322–658)	498 (327–617)	0.9
Coincident diagnosis	21 (46.6)	27 (26.4)	0.02
First episode	35 (77.7)	75 (73.5)	0.6
Clinical presentation
Asymptomatic	6 (13.3)	31 (30.3)	0.03
Chancre	8 (17.7)	15 (14.7)	0.6
Secondary syphilis	28 (62.2)	46 (45.0)	0.07
Other	3 (6.6)	8 (7.8)	1.0
RPR≥1/64	20 (44.4)	25 (24.5)	0.007

LI = liver involvement; HBV/HCV = hepatitis B virus/hepatitis C virus; ART = antiretroviral therapy; VL = viral load; RPR = rapid plasma reagin

Quantitative variables: mean (IQR). Qualitative variables: n (%)

*Of those with prior HIV diagnosis

^†Of those on ART

Figures 1–4 show the course of the liver enzymes (ALT, alkaline phosphatase, GGT and bilirubin) of all 45 patients with liver involvement; note the rise in these enzymes coinciding with the diagnosis of syphilis and their subsequent fall.

Figure 1

Course of alanine transaminase (UI/mL) in all patients with liver involvement

Figure 2

Course of alkaline phosphatase (UI/mL) in all patients with liver involvement

Figure 3

Course of gamma-glutamyltransferase (UI/mL) in all patients with liver involvement

Figure 4

Course of bilirubin (mg/dL) in all patients with liver involvement

DISCUSSION

The results of this study show that one-third of HIV-infected patients diagnosed with early syphilis had liver involvement, with the only factor associated with this event being an RPR titre ≥1/64.

Although Paracelsus described liver involvement due to syphilis in the general population in 1585, only a few data have actually been published, with the largest recent series comprising 17 cases.¹² Liver involvement has traditionally been described as a disproportionate increase in alkaline phosphatase in the context of secondary syphilis with rash.^13,14 In HIV-infected patients it was not until the last decade that data appeared about liver involvement in syphilis.^4–8 The two largest series include seven and 12 cases, respectively,^4,8 reporting a prevalence of liver involvement in HIV-infected patients with early syphilis of around 40%.⁸ As in our study, liver involvement due to syphilis in these series is similar to that described for the general population, with a marked increase in alkaline phosphatase and a slightly lower rise in transaminases, with generalized rash being the most common associated symptom.^4,13,14 In our series, 51.9% of patients presented with secondary syphilis with generalized rash, with no difference between those with liver involvement and those without. Of note too, was the fact that syphilis was asymptomatic in over one-third of our cases, with the diagnosis made from routine serological tests.

Concerning the factors associated with liver involvement in HIV-infected patients with early syphilis, Mullick et al. ⁴ found a correlation with a higher CD4 cell count, suggesting that host immune inflammatory response in the periportal region might be a factor in the development of clinical symptoms of syphilitic hepatitis. In our series, the immunological status was similar in patients with and without liver involvement, probably because a considerable percentage of our cases occurred in patients under regular follow-up and receiving cART, with good immunovirological status. Crum-Cianflone et al. ⁸ reported a greater likelihood of syphilitic hepatitis in patients who had had HIV infection for longer. No association was found in our series with the duration of HIV infection, the CD4 cell count, chronic co-infection with hepatotropic viruses or with the clinical presentation of syphilis (asymptomatic, chancre or secondary). Like Mullick et al.,⁴ we also saw a correlation with high RPR titres, which may represent a higher treponemal load with potential risk of organ involvement. As seen in other series,^4,8 after standard treatment of syphilis the liver enzyme levels returned to baseline values in all cases.

To date this is the largest reported series of HIV-infected patients with early syphilis and liver involvement. Based on our findings, syphilis should be included in the differential diagnosis of liver enzyme elevation in HIV-infected patients and, as indicated in the guidelines, patients should be screened for syphilis on diagnosis of their HIV infection as well as periodically afterwards and preventive measures encouraged.

Conflict of interest: None declared.

References

Poles

, Lew

, Dieterich

. Diagnosis and treatment of hepatic disease in patients with HIV. Gastroenterol Clin North Am 1997;26:291–321

Ogedegbe

, Sulkowski

. Antiretroviral-associated liver injury. Clin Liver Dis 2003;7:475–99

Wnuk

. Liver damage in HIV-infected patients. Med Sci Monit 2001;7:729–36

Mullick

, Liappis

, Benator

, Roberts

, Parenti

, Simon

. Syphilitic hepatitis in HIV-infected patients: a report of 7 cases and review of the literature. Clin Infect Dis 2004;39:100–5

De Miguel

, Benedicto

. Hepatitis sifilítica: una patología ‘poco frecuente’. Mediafam 2002;3:226–8

Cooper

, MacPherson

, Angel

. Liver toxicity resulting from syphilis and Jarish-Herxheimer reaction in cases of coinfection with HIV and hepatitis C virus. Clin Infect Dis 2005;40:1211–2

Albandea Moreno

, Aguilar Urbano

, Rivera Irigoin

, Syphilitic hepatitis: case report. Rev Esp Enferm Dig 2009;101:813–9

Crum-Cianflone

, Weekes

, Bavaro

. Syphilitic hepatitis among HIV-infected patients. Int J STD AIDS 2009;20:278–84

Dooley

, Tomski

. Syphilitic pneumonitis in an HIV-infected patient. Chest 1994;105:629–31

10.

Sabbatani

, Manfredi

, Attard

, Salfi

, Chiodo

. Secondary syphilis presenting with acute severe hepatic involvement in a patient with undiagnosed HIV disease. AIDS Patient Care STD 2005;19:545–9

11.

Adachi

, Koibuchi

, Okame

, Case of secondary syphilis presenting with unusual complications: syphilitic proctitis, gastritis, and hepatitis. J Clin Microbiol 2011;49:4394–6

12.

Feher

, Somogyi

, Timmer

, Józsa

. Early syphilitic hepatitis. Lancet 1975;2:896–9

13.

Baker

, Kaplan

, Wolfe

, McGowan

. Liver disease associated with early syphilis. N Engl J Med 1971;284:1422–3

14.

Campisi

, Whitcomb

. Liver disease in early syphilis. Arch Intern Med 1979;139:365–6