Prevalence of hepatitis in early syphilis among an HIV cohort

Abstract

To investigate the prevalence of syphilitic hepatitis among a group of HIV-infected patients we performed a cross-sectional observational study of consecutive HIV-infected patients with early syphilis attending University Hospital Birmingham between 1 January 2005 and 31 August 2008. The AIDS Clinical Trials Group grading for abnormal liver enzymes was used to identify hepatitis. A total of 62 HIV-infected patients were diagnosed with early syphilis during the study period. Twelve (19.3%) of them demonstrated abnormal liver enzymes consistent with syphilitic hepatitis involving raised levels of alanine aminotransferase, aspartate transaminase, alkaline phosphatase or gamma-glutamyl transferase (GGT). Grade 3 hepatotoxicity was observed among five patients. None of the patients with syphilitic hepatitis had grade IV hepatitis or abnormal bilirubin levels. Liver biopsy was not carried out in any of the patients, and following completion of treatment of syphilis all abnormal liver enzymes returned to normal levels after a median of 16 weeks. Exclusion of syphilis must be considered when investigating hepatic disease in HIV-infected patients.

Keywords

HIV syphilis hepatitis prognosis prevalence

BACKGROUND

Abnormality of the liver enzyme profile of HIV-infected patients can be the result of several significant infectious causes. Hepatitis secondary to infection with Treponema pallidum is an uncommon presentation of early and early latent syphilis in HIV-uninfected patients.^1,2 A significant proportion of HIV-infected patients are at risk of T. pallidum infection.^3,4 It is probable that some of those patients may present with the less common complications of syphilis, including hepatitis. Few data on the prevalence of syphilitic hepatitis among HIV-infected patients are available. The aim of the present study was to investigate the prevalence of syphilitic hepatitis among a group of HIV-infected patients diagnosed consecutively with early syphilis.

METHODS

Study population and case definition

This was a cross-sectional observational study of consecutive HIV-infected patients with early syphilis attending University Hospital Birmingham between 1 January 2005 and 31 August 2008. In our department, syphilis serology is routinely requested for HIV-infected patients on three-monthly basis. Consequently we have been able to diagnose most infected cases at the early stage of syphilis.

Laboratory procedures

Serum samples of patients with positive enzyme immunoassay (EIA) for treponemal IgG antibodies were re-tested for confirmation with venereal disease research laboratories (VDRL), T. pallidum particle agglutination assay (TPPA) and EIA for treponemal IgM antibodies. Specimens testing positive for EIA-IgG and TPPA were considered as indicative of syphilis. Four-fold reductions of VDRL titre after treatment were considered as indicative of successful treatment.

Liver function tests (LFTs) prior to a diagnosis of early syphilis, at the time of diagnosis and after a four-fold drop in VDRL, were documented. These were alanine transaminase (ALT) (reference interval 5–41 U/L), aspartate aminotransferase (AST) (reference interval 5–43 U/L), alkaline phosphatase (ALP) (reference interval 70–320 U/L) gamma-glutamyl transferase (GGT) (reference interval 9–50 U/L) and bilirubin (Bil) (reference interval 1–22 μmol/L).

Abnormalities in LFTs were graded according to AIDS Clinical Trials Group (ACTG) criteria: grade 1 (1.25–2.5 × upper limit of normal [ULN]), grade 2 (2.5–5.0 × ULN), grade 3 (5–10 × ULN) and grade 4 (>10 × ULN)⁵.

Case definition

Syphilitic hepatitis was defined as abnormality of equal or greater than 1.25 times ULN (grade 1 or above) in any of the LFTs within a 90-day period prior to the diagnosis of syphilis, in the absence of an alternative cause for hepatitis that resolved with treatment of syphilis.

Cases with deranged liver enzymes of more than 90 days prior to diagnosis of early syphilis, those with co-infection with hepatitis A, B or C, acute toxoplasmosis, acute Epstein–Barr virus or cytomegalovirus at the time of abnormal LFTs were not considered to have syphilitic hepatitis in this study. In addition, patients with a history of alcohol abuse or drug-induced hepatitis were excluded.

In order to exclude acute infections with hepatitis B (HBV) and hepatitis C (HCV), assays for HBV DNA and HCV RNA were requested in all patients.

Data collection

Measurements of CD4 T-lymphocyte count and plasma HIV viral load at the time of diagnosis of syphilis, risk factor for HIV acquisition, dates and type of treatment given for syphilis, history of being on combination antiretroviral therapy, and the results of other investigations performed to uncover the cause of hepatitis were documented.

Results were recorded on a Microsoft Excel spreadsheet. Median values with interquartile (IQR) ranges were calculated. The Mann–Whitney test was used to investigate the difference between non-parametric data. Fisher's exact test was used for categorical data.

RESULTS

A total of 62 consecutive HIV-infected patients were diagnosed with early syphilis during the study period. Twelve (19.3%) of them demonstrated abnormal liver enzymes consistent with syphilitic hepatitis. Table 1 summarizes the demography of patients with syphilitic hepatitis compared with that of patients without syphilitic hepatitis. None of the investigated factors were significantly associated with syphilitic hepatitis.

Table 1

Demographics of study population at the time of diagnosis of early syphilis

	HIV and syphilis (n = 50)	HIV and syphilitic hepatitis (n = 12)	P value
Median age (IQR), years	36 (31–42.75)	40 (35.5–46.5)	0.209
Men	49	12
Women	1	0
Risk factor for HIV
MSM	42 (84%)	11 (92%)	0.82
Heterosexual	5 (10%)	0
Bisexual	1 (2%)	1 (8%)
Unknown	2 (4%)	0
Median VDRL	1:16 (1:8–1:32)	1:32 (1:16–1:64)	0.241
Median CD4 count (IQR), cells/mm³	420 (300–545)	388 (294–479)	0.260
Log₁₀ median HIV viral load (IQR), copies/mL	4.1 (1.8–4.8)	5.0 (3.2, 5.7)	0.09
On combination antiretroviral therapy	17 (34%)	5 (41%)	0.87
Syphilis treatment with benzathine penicillin*	43 (86%)	10 (83%)	0.055

IQR=interquartile range; MSM= men who have sex with men; VDRL= Venereal Diseases Research Laboratory

A value of 49 was considered for viral load of less than 50 copies/mL

*Benzathine penicillin 2.4 mU/im weekly for three weeks. The remaining patients were treated with doxycycline 100 mg twice daily for 14 days

Seven patients without and none of the patients with syphilitic hepatitis had CD4 count of less than 200 cells/mm³.

Of the 12 patients with syphilitic hepatitis, nine had physical signs: seven displayed generalized or localized maculopapular rash. Other signs present were alopecia, weight loss, myalgia, malaise, headaches, fever, night sweats and right hypochondrial pain. Three patients were asymptomatic. There was no relationship between raised LFTs and any of the documented signs of syphilis.

Syphilitic hepatitis was associated with raised levels of ALT, AST, ALP or GGT (Table 2). Grade 3 hepatotoxicity was recorded among five patients; two with raised ALP, two with raised GGT and one patient with raised ALT. None of the patients with syphilitic hepatitis had grade IV hepatitis or abnormal Bil levels. Liver biopsy was not carried out in any of the patients.

Table 2

Summary of biochemical markers of liver function observed in patients with syphilitic hepatitis (n = 12)

	Alanine transaminase (U/L)	Aspartate aminotransferase (U/L)	Alkaline phosphatase (U/L)	Gamma-glutamyl transferase (U/L)	Bilirubin (μmol/L)
Median value (IQR)	52 (47, 77)	45 (31, 68)	351 (247, 725)	1.0 (0.83, 2.5)	8 (6,13)
Normal range	5–41	5–43	70–320	9–50	1–22
Total number of patients with abnormal test*	5 (71%)^†	5 (36%)	5 (36%)	5 (36%)	0

IQR=interquartile range

*Elevations above 1.25 times upper limit of normal range

^†Percentage of the total of seven patients with alanine transaminase assay alanine transaminase was not requested in five patients

^‡Gamma-glutamyl transferase was not requested in eight patients

All episodes of syphilis achieved four-fold or more decline in VDRL titre after treatment. Following completion of treatment of syphilis all abnormal LFTs returned to levels within the reference range after a median of 16 (IQR 11, 20) weeks.

None of the treated patients with syphilitic hepatitis reported clinical symptoms of Jarisch–Herxheimer reaction.

DISCUSSION

Syphilitic hepatitis is uncommon among HIV-uninfected patients. The methodologies used in previous reports have led to wide difference in the prevalence of this complication of syphilis; between 0.2%¹ and 10%.⁶ Most of those studies considered any elevation beyond ULN in the presence of syphilis as hepatitis. In the largest study among 175 patients with syphilis, 17 had histologically confirmed syphilitic hepatitis².

The definition of syphilitic hepatitis in the present study was more stringent than previous reports. The prevalence of syphilitic hepatitis among HIV uninfected patients with the criteria used in our study may be less than that already reported. It should be noted that ACTG grading is not the only scaling system for abnormal liver enzymes. Another definition uses mild, moderate and severe grades to scale abnormal liver enzymes;⁷ because that system also considers levels above 1.25 × ULN as abnormal we do not believe the use of an alternative grading system in the present study would have changed the prevalence of syphilitic hepatitis.

In the present study, the prevalence of syphilitic hepatitis among a group of HIV-infected patients diagnosed consecutively with early syphilis was 19.3%. The surprisingly high prevalence of syphilitic hepatitis highlights the need for exclusion of syphilis as a cause of acute hepatitis among HIV-infected patients.

The only other report on syphilitic hepatitis in HIV-infected patients is based on findings in seven cases where ALP and Bil were the most likely LFT to be elevated.⁸ In the present study ALT was raised in higher proportion than other LFTs. ALP was raised in the same proportion of patients as were AST and GGT while none of the patients had hyperbilirubinaemia. The difference between our findings and that of the cases reported by Mullick et al. may be have been due to the detection of syphilitic hepatitis at an earlier stage in our cohort. It is likely that syphilitic hepatitis may present as hepatocellular-predominant liver injury during early syphilis that may evolve into cholestasis at later stages of syphilis if untreated. Alternatively, it may be likely that syphilitic hepatitis may present as a mixed pattern of liver injury. Unlike Mullick et al., we could not find an association between syphilitic hepatitis with CD4 count or VDRL titre.

In our study grade 3 hepatotoxicity was recorded in five (41%) patients with syphilitic hepatitis; fortunately none of them developed grade 4 hepatotoxicity.

Syphilitic hepatitis can develop into severe hepatic failure despite treatment of syphilis; two such cases diagnosed during secondary stage of syphilis have been reported in the literature.^9,10 As syphilitic hepatitis may result in widespread hepatic necrosis, even after treatment, we believe monitoring of patients’ LFTs after treatment of syphilis may be justified, especially for patients diagnosed after the primary stage of syphilis.

Syphilitic hepatitis has been reported to be reversible in the majority of reported cases among non-HIV-infected patients.^11–13 The present study suggests that syphilitic hepatitis in HIV-infected patients diagnosed in early stages of infection may have similar prognosis as all cases resolved after treatment of syphilis. Raised LFTs returned to their reference range within a median of 16 weeks after treatment of syphilis.

In our cohort, doxycycline seemed to have been used in more patients with syphilitic hepatitis; this was mainly due to patients’ choice to avoid intramuscular injections. We believe the antibiotic of choice for treatment of syphilitic hepatitis is penicillin.

All patients with syphilitic hepatitis in our study were men who have sex with men (MSM). Previous case reports of syphilitic hepatitis in HIV mostly describe male patients,^2,3,4 and anal sex has been proposed as risk factor for syphilitic hepatitis on that ground. It has been hypothesized that through direct access of treponemes to the liver via the portal circulation after anorectal infection syphilitic hepatitis may develop.¹⁴ Few data exist to support this hypothesis as cases of syphilitic hepatitis in heterosexual men have also been reported.^12,13,15 Higher rates of syphilitic hepatitis among MSM may be a function of the higher prevalence of syphilis in this group of patients.¹⁶

To our knowledge the present study is the only available report on the prevalence of syphilitic hepatitis among HIV-infected patients. Exclusion of syphilis must be considered as part of the investigations for hepatic disease in HIV-infected patients.

Footnotes

Results presented in the British HIV Association Spring Conference, 2008.

Conflict of interest: None to declare.

Funding resources: None to declare.

References

Hahn

. Syphilis of the liver. Am J Syph, Gonorrhoea, Vener Dis 1943;27:529–62

Feher

, Somogyi

, Timmer

, Jozsa

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

Lynn

, Lightman

. Syphilis and HIV: a dangerous combination. Lancet Infect Dis 2004;4:456–66

Frauenfelder

. Incidence of syphilis in UK rises as HIV diagnoses hold steady. BMJ 2006;333:1089

Adult AIDS Clinical Trial Group (AACTG). Table for Grading Severity of Adult Adverse Experiences. Bethesda, MD: National Institute of Allergy and Infectious Diseases, 1992

Jorsa

, Timmer

, Somogyi

, Feher

. Hepatitis syphilitica. A clinico-pathological study of 25 cases. Acta Hepatogastroenterol 1977;24:344–7

Giannini

, Testa

, Savarino

. Liver enzyme alteration: a guide for clinicians. CMAJ 2005;172:367–79

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;e100–5

, Harrison

, Hunter

. Syphilitic hepatitis resulting in fulminant hepatic failure requiring liver transplantation . J Infect 2007;54:e115–7

10.

Sabbatani

, Manfredi

, Attard

, Salfi

, Chiodo

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

11.

Ishikawa

, Shimizu

, Uehara

, A patient with early syphilis complicated by fatty liver who showed an alleviation of hepatopathy accompanied by jaundice after receiving anti-syphilitic therapy. Intern Med 2006;45:953–6

12.

Tang

, Thin

RNT

, Croft

. Diagnosis Nephrotic syndrome and hepatitis in early syphilis. Postgrad Med J 1989;65:14–5

13.

Sehgal

, Rege

. Malignant syphilis and hepatitis Case report. Br J Vener Dis 1974;50:237–8

14.

Campisi

, Whitcomb

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

15.

Mandachel

, Cocal

, Caro-Samparal

, A forgotten aetiology of acute hepatitis in immunocompetent patient: syphilitic infection. J Intern Med 2006;259:214–5

16.

Health protection agency. All new STI episodes seen at GUM clinics in the UK: 1998–2007. See http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/1215589014474 (last checked 16 July 2012)