If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Dr. Arturo González Quintela, Department of Internal Medicine Hospital Clínico Universitario, 15706 University of Santiago de Compostela, Spain.
Liver involvement during infectious mononucleosis is common, but jaundice is considered rare. This study aimed to investigate serum bilirubin concentrations in patients with infectious mononucleosis and immune abnormalities associated with jaundice.
Methods
We report on an adult patient with monoclonal B lymphocytosis and IgM-lambda gammopathy who developed a severe icteric hepatitis during infectious mononucleosis. We then reviewed the clinical records of 389 patients admitted to the hospital with infectious mononucleosis between 1995 and 2018 (51.7% male patients; median age, 19 years; range, 15-87 years) with focus on liver abnormalities and associated factors.
Results
Fifty-nine patients (15.1%) had serum bilirubin concentrations between 1.5 and 3 mg/dL, and 47 patients (12.0%) had serum bilirubin >3 mg/dL. Patients with increased bilirubin concentrations had a distinct clinical presentation, with more frequent abdominal pain, nausea and vomiting, and less frequent sore throat than patients with normal bilirubin. Age and sex were not significantly different for the patients with increased and normal serum bilirubin concentrations. The patients with increased serum bilirubin concentrations showed higher levels of immune activation markers than the patients with normal bilirubin, including blood lymphocyte counts, serum IgM, and β2-microglobulin concentrations. Heterophile antibody-positive patients (88.6%) showed similar bilirubin concentrations but higher aspartate aminotransferase and alkaline phosphatase levels than their heterophile-negative counterparts. Serum bilirubin elevations normalized quickly during follow-up.
Conclusions
Transient hyperbilirubinemia is common during severe (in-hospital) infectious mononucleosis in adult patients. Patients with hyperbilirubinemia have less frequent pharyngitis symptoms and more frequent abdominal symptoms. Hyperbilirubinemia during infectious mononucleosis is associated with immune activation markers.
Infectious mononucleosis produced by primary Epstein-Barr virus (EBV) infection is characterized by fever, pharyngotonsillitis, lymphadenopathy, blood lymphocytosis, and a variety of potential systemic complications.
Liver involvement in the form of acute hepatitis (up to 5-fold increase in serum transaminase levels) is nearly universal in patients with infectious mononucleosis
and is usually mild, clinically undetected, and resolves spontaneously. Jaundice is distinctly uncommon. Only 24 cases of cholestatic hepatitis were found in an extensive literature review in 2005.
Isolated cases of severe icteric hepatitis during infectious mononucleosis have been published in recent years, which is a sign of the disorder's rarity.
The pathogenesis of severe liver injury during infectious mononucleosis is unclear, but might be immune-mediated. The classical description of liver histopathology during infectious mononucleosis includes a characteristic mononuclear infiltration in lobular sinusoids in relation to liver injury markers.
In this article, we report on an adult patient with a low-grade lymphoproliferative disorder (monoclonal B lymphocytosis) who developed severe icteric hepatitis during EBV-induced infectious mononucleosis. We subsequently studied serum bilirubin concentrations in a larger sample of adult (older than 15 years) patients with infectious mononucleosis and factors associated with jaundice, particularly immune abnormalities.
Methods
Design and setting
An index case with severe, icteric hepatitis during infectious mononucleosis (see below) prompted the review of the clinical records of adult (15-years or older) patients with infectious mononucleosis who were admitted to the Department of Internal Medicine of the Santiago de Compostela (Spain) University Hospital between 1995 and 2018, as reported elsewhere.
The hospital is a reference center for an area of approximately 400,000 inhabitants. The main reasons for hospital admission were severe signs of systemic inflammatory response, difficulty with oral intake, and the presence of complications.
Ethical approval
The study was reviewed and approved by the institutional Ethics Committee (code 2017/578) who waived the requirement for informed consent from the study participants, in agreement with Spanish regulations for retrospective studies of clinical records. Written informed consent was obtained from the detailed case report patient.
Diagnostic criteria for infectious mononucleosis
A definitive diagnosis of infectious mononucleosis was considered when the clinical syndrome was accompanied by positive immunoglobulin (Ig)-M antibodies against the viral capsid antigen (VCA) of EBV and/or a positive heterophile antibody result, as revealed by either a classic Paul-Bunnell test or a commercial rapid test.
A total of 401 patients met the diagnostic criteria; however, baseline (acute phase) serum bilirubin measurements were not available for 12 cases. The study therefore included 389 patients (51.7% male patients; median age, 19 years; range, 15-87 years). We reviewed their clinical records, with a special focus on serum bilirubin, markers of liver damage (serum aspartate aminotransferase [AST], alanine aminotransferase [ALT], gamma-glutamyltransferase [GGT], alkaline phosphatase [AP], and tissue polypeptide specific antigen [TPS, keratin-18 fragment]), and immune activation markers (blood leukocyte count, lymphocyte count, and serum concentrations of IgG, IgA, IgM, and β2-microglobulin). The first available determination, usually taken on admission, was used for analyses. Given that the methods for determining liver markers and their reference values have changed over the years, the results are expressed as the ratio of the upper limit of normal (ULN) to the measured values. Reference values for TPS were 0-100 U/L.
Reference values for IgG, IgA, IgM, and β2-microglobulin were 700–1600 mg/dL, 70–400 mg/dL, 40-230 mg/dL, and 1.09-2.53 mg/L, respectively. In addition to the case reported below, who had a monoclonal B-cell lymphocytosis which was probably present before the mononucleosis episode, only 6 other patients had some type of immune disorder prior to EBV infection due to pre-existing diseases or their treatment (three patients with inflammatory bowel disease, one patient with multiple sclerosis, one patient with cutaneous lupus, and one patient with generalized psoriasis). We also reviewed the clinical records of the patients with baseline bilirubin concentrations >3 mg/dL for follow-up to bilirubin normalization.
Statistical analyses
We employed the Mann-Whitney test to compare the numerical data between groups and the Wilcoxon test to compare paired samples of numerical values. We employed the chi-squared test (with continuity correction and trend analysis, when appropriate) to compare proportions and the Jonckheere-Terpstra test for trend analysis of numerical variables among ordinal categories. Lastly, we used the Spearman's rank test to assess the correlations. All tests were two-tailed.
Results
Case presentation
A 48-year-old man with no remarkable previous disease was admitted to the hospital in 2016 due to fever, vomiting, and liver abnormalities. His symptoms had started 5 days earlier with general malaise and chills. He reported no sore throat or any other symptoms. The physical examination revealed jaundice, fever, and the presence of small cervical lymph nodes. The main findings from the complementary examinations were the presence of lymphocytosis in peripheral blood, with abundant (> 15%) atypical forms (activated lymphocytes), hypertriglyceridemia (297 mg/dL), elevated serum liver enzymes (AST 250 IU/L, ALT 374 IU/L, GGT 521 IU/L, AP 387 IU/L), and hyperbilirubinemia (4.6 mg/dL). The abdominal ultrasound and CT scan highlighted liver enlargement, splenomegaly (20 cm), multiple small lymph nodes in all regions, and signs of right colitis. The serological tests for human immunodeficiency virus, hepatitis B virus, and hepatitis C virus were negative. The tests for hepatitis A virus, herpes simplex virus, varicella-zoster virus, and cytomegalovirus revealed only past infection (presence of positive IgG with negative IgM). The serum heterophile antibody test was negative; however, the IgM against the EBV-VCA was positive, and EBV-DNA was detected in serum (2700 copies/mL). Serum antinuclear antibodies were negative. Serum concentrations of IgG, IgA, and IgM were 875 mg/dL, 184 mg/dL, and 766 mg/dL (maximum level during admission, 1340 mg/dL), respectively. Serum protein electrophoresis and immunofixation revealed IgM-lambda monoclonal gammopathy. Serum β2-microglobulin was 9.87 mg/L. During the first week of hospitalization under symptomatic therapy, the patient developed watery diarrhea and maintained fever, lymphocytosis, and elevated transaminase levels; the hyperbilirubinemia worsened and reached a maximum of 19.0 mg/dL (Fig. 1). The prothrombin time was within normal limits. A liver biopsy on the seventh day showed intense portal hepatitis with interface activity, centrilobular necrosis, and marked inflammatory sinusoidal infiltration by CD3-positive lymphocytes, compatible with EBV-induced hepatitis (Fig. 1). Signs of lymphomatous infiltration were absent on both the liver histology and flow cytometry. Intravenous acyclovir and methyl-prednisolone therapy was initiated and was followed by a progressive improvement in symptoms and laboratory data (Fig. 1). Broad-spectrum antibiotics were also added due to an episode of Gram-negative bacteremia. The patient was discharged on the eighteenth day.
Fig. 1Upper panel: Main biological determinations, period of drug administration, and time of liver biopsy and initial Epstein-Barr virus studies during acute infectious mononucleosis in the reported case. Acyclovir was administered intravenously (600 mg every 8 h); methyl-prednisolone was administered intravenously (60 mg daily) and subsequently switched to the oral route in tapering doses. An IgM-lambda monoclonal gammopathy was detected on a serum sample taken of the day of hospital admission and remained detectable during the following years. Serum total IgM and β2-microglobulin on admission were 766 mg/dL and 9.87 mg/L, respectively. Serum IgM reached a maximum of 1340 mg/dL on the 16th day. Previous and subsequent biological determinations (in the preceding and following years, respectively) are also shown. Serum AST is represented as the fold increase in the upper limit of normal (63 IU/L); the total blood lymphocyte count is the sum of the normal and atypical (activated) lymphocytes, as measured in the standard counter. Lower panels: Representative images of liver biopsy. (A) The hepatic tissue shows a predominantly lobular pattern of inflammation with numerous intrasinusoidal lymphocytes (hematoxylin & eosin; original magnification 10x); (B) intrasinusoidal lymphocytes are more evident at higher magnification (hematoxylin & eosin; original magnification, 40x); (C) immunohistochemistry with CD3 highlights T cells forming lines within the sinusoids (original magnification, 40x). EBV, Epstein-Barr virus.
Serum bilirubin and liver enzymes normalized during the following weeks and remained within normal limits in the following 5 years (Fig. 1). The IgM-lambda monoclonal gammopathy persisted, although total serum IgM concentrations lowered to 270-290 mg/dL in successive checkups in the following years. Serum β2-microglobulin also progressively descended in the months after admission, but remained between 2.93 mg/L and 3.63 mg/L in the following years. The IgM against EBV-VCA disappeared during follow-up, and the patient developed IgG antibodies against EBV-VCA and against EBV nuclear antigen (EBNA), further confirming the EBV primary infection during the acute episode. Mild lymphocytosis (4.10-6.10 × 109 cells/L) persisted during follow-up (Fig. 1). The peripheral blood immunophenotype revealed a monoclonal B-cell population of chronic lymphocytic leukemia (CLL) type, composed of CD5(+ve), CD19(+ve), CD23(+ve), CD20(+ve), and surface IgM(+ve) cells with lambda expression. The amount of this clonal population in peripheral blood remained between 1.03 and 1.52 × 109 cells/L, which is compatible with the diagnosis of monoclonal B-cell lymphocytosis with IgM-lambda gammopathy but not fulfilling the criteria for CLL. A retrospective review of blood tests from this patient revealed that mild lymphocytosis (5.80 × 109 cells/L) was present at least one year before the acute infectious mononucleosis episode (Fig. 1). No more significant infections had developed either before or after the episode. Five years later, the patient remains completely asymptomatic and under no specific therapy.
Case series
Fig. 2 presents a histogram of the serum bilirubin concentrations for the patients with infectious mononucleosis. The median serum bilirubin concentration was 0.8 mg/dL (range, 0.1–32.8 mg/dL; interquartile range [IQR], 0.5–1.70 mg/dL). Serum bilirubin was within normal limits in most cases (Fig. 2). Table 1 shows a comparison of the demographic, clinical, and biochemical characteristics of the patients with normal (<1.5 mg/dL) bilirubin concentrations, those with serum bilirubin concentrations of 1.5–3 mg/dL (n = 59, 15.1%), and those with serum bilirubin concentrations >3 mg/dL (n = 47, 12.0%). Thirty patients (7.7%) had serum bilirubin concentrations ≥5 mg/dL, and 5 patients (1.2%) had serum bilirubin concentrations ≥10 mg/dL. The median percentage of direct bilirubin from total bilirubin was 77.6% (IQR, 66.4%–82.7%) in 95 cases with elevated total bilirubin concentrations (≥1.5 mg/dL) and available differential determination.
Fig. 2Histogram of serum bilirubin concentrations on admission in the study population. Cases with bilirubin concentrations >11 mg/dL (n = 2) are not represented. These corresponded to a previously healthy 26-year old male with a serum bilirubin of 32.8 mg/dL, and a 21-year old male with a previous diagnosis of Gilbert syndrome and a serum bilirubin of 24.4 mg/dL on admission.
Table 1Demographic, clinical, and biological characteristics of patients with infectious mononucleosis, stratified by baseline serum bilirubin concentration.
Serum bilirubin
<1.5 mg/dL
1.5-3 mg/dL
>3 mg/dL
P-value
No.
No.
No.
Age (years)
283
19 (17-23)
59
18 (17-24)
47
20 (17-25)
0.697
Sex (male)
283
140 (49.5)
59
34 (57.6)
47
27 (57.4)
0.191
Sore throat
281
209 (74.4)
59
42 (71.2)
47
26 (55.3)
0.012
Malaise
282
197 (69.9)
59
48 (81.4)
47
39 (83.0)
0.021
Abdominal pain
282
68 (24.1)
59
23 (39.0)
47
17 (36.2)
0.019
Nausea or vomiting
281
57 (20.3)
59
17 (28.8)
47
24 (51.5)
<0.001
Lymphadenopathy
281
211 (75.1)
59
44 (74.6)
47
34 (72.3)
0.704
Previous corticosteroid therapy
281
24 (8.5)
57
1 (1.8)
47
0 (0.0)
0.009
In-hospital corticosteroid therapy
280
107 (38.2)
59
18 (30.5)
47
13 (27.7)
0.102
Hospital stay (days)
283
5 (4-7)
59
6 (4-8)
47
7 (5-10)
<0.001
Positive heterophile antibody test
267
236 (88.4)
56
53 (94.6)
46
38 (82.6)
0.614
Serum AST (IU/L)
282
85 (41-167)
59
159 (87-239)
47
236 (156-344)
<0.001
Serum AST (times the ULN)
282
3.24 (1.56-6.08)
59
5.60 (3.48-9.04)
47
7.43 (6.10-12.0)
<0.001
Serum ALT (IU/L)
282
169 (77-299)
59
284 (202-448)
47
308 (232-518)
<0.001
Serum ALT (times the ULN)
282
5.59 (2.37-9.48)
59
9.31 (6.00-13.6)
47
9.41 (7.39-13.7)
<0.001
Serum GGT (IU/L)
282
94 (43-159)
59
179 (118-275)
47
229 (134-359)
<0.001
Serum GGT (times the ULN)
282
2.29 (1.10-3.68)
59
4.42 (3.09-5.86)
47
4.96 (3.22-7.11)
<0.001
Serum AP (IU/L)
277
293 (171-448)
59
520 (379-636)
45
537 (396-866)
<0.001
Serum AP (times the ULN)
277
1.43 (0.81-2.19)
59
2.67 (1.94-3.45)
45
2.87 (2.05-4.05)
<0.001
Serum TPS (U/L)
12
192 (76-444)
4
615 (568-900)
8
626 (526-1590)
0.009
Blood leukocytes (x109/L)
283
11.4 (7.16-15.8)
59
13.7 (10.1-16.8)
47
11.5 (7.76-17.7)
0.043
Blood lymphocytes (%)
283
50.9 (40.4-59.9)
59
56.6 (44.6-64.2)
47
57.3 (47.7-64.3)
<0.001
Serum IgG (mg/dL)
177
1300 (1080-1520)
39
1370 (1095-1660)
41
1440 (1110-1680)
0.176
Serum IgA (mg/dL)
177
274 (186-354)
39
300 (197-376)
41
280 (202-365)
0.551
Serum IgM (mg/dL)
177
226 (169-287)
39
271 (196-350)
41
298 (215-411)
<0.001
Serum β2-microglobulin (mg/L)
69
4.14 (3.52-5.52)
16
5.28 (4.18-6.78)
18
6.00 (4.67-7.02)
<0.001
Data are medians and interquartile ranges (within parentheses) or absolute numbers and percentages (within parentheses). No., number of patients with available determination. P-values were obtained with a trend test (chi-square test for categorical variables and Jonckheere-Terpstra test for numerical variables). AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; AP, alkaline phosphatase; ULN, upper limit of normal; TPS, tissue polypeptide specific antigen (keratin-18 fragment; normal reference values 0-100 U/L).
Two patients with bilirubin concentrations of 24.4 mg/dL (direct fraction, 15.4 mg/dL) and 2.7 mg/dL (direct fraction, 1.3 mg/dL) had a previous diagnosis of Gilbert's syndrome. Only one patient (serum bilirubin, 4.5 mg/dL) had a history of alcohol abuse. No other significant cause of chronic liver disease was present in the clinical records of the remaining patients.
Serum transaminase levels were elevated in the vast majority of cases. The median AST level was 4.0 times the ULN (IQR, 1.9-7.7 times, n = 388), the median ALT level was 6.7 times the ULN (IQR 3.0-10.7 times, n = 388), the median GGT level was 2.8 times the ULN (IQR, 1.3-4.8 times, n = 388), and the median AP level was 1.8 times the ULN (IQR, 1.0-2.6 times, n = 381). Serum transaminase concentrations were correlated with those of serum bilirubin (Table 1). Serum concentrations of tissue polypeptide specific antigen (TPS, keratin-18 fragment) were also higher in the patients with increased bilirubin concentrations (Table 1).
The patients with increased serum bilirubin concentrations had a clinical presentation that was, as a whole, different from that of the patients with normal bilirubin concentrations (Table 1). Malaise and abdominal symptoms (pain, nausea, and vomiting) tended to be more frequent among the patients with increased bilirubin concentrations. Conversely, typical symptoms such as sore throat were less common in patients with increased bilirubin concentrations (Table 1).
A total of 25 patients (6.5% of those with available data) received corticosteroid therapy prior to admission. Short-term corticosteroids were generally indicated by the general practitioner due to intense oropharyngeal symptoms that made swallowing difficult. The corticosteroid was methyl-prednisolone in 9 cases, deflazacort in 5 cases, prednisone in 4 cases, and unknown in 7 cases. The exact duration and dose of previous corticosteroid therapy was not recorded. Such previous corticosteroid therapy was negatively associated with bilirubin concentrations on hospital admission (Table 1).
Sex was not significantly associated with abnormal bilirubin concentrations (Table 1). However, median bilirubin concentrations were slightly higher in males than in females (Table 2). Male patients were older and showed higher white blood cell counts than females. Females reported nausea or vomiting more frequently than males and showed higher serum IgM concentrations (Table 2). Age was not significantly associated with abnormal bilirubin concentrations (Table 1). Younger patients (<25 years) presented typical symptoms of mononucleosis (sore throat and lymphadenopathy) more often than older patients (Table 2). Younger patients also had higher white blood cell counts, higher lymphocyte counts, and higher serum IgG and IgM concentrations (Table 2). Neither sex nor age were significantly associated with levels of liver injury markers such as AST, ALT, GGT, and AP (Table 2).
Table 2Clinical and biological characteristics of patients with infectious mononucleosis, stratified by sex and age.
Sex
Age
Female
Male
<25 years
≥25 years
No.
No.
P-value
No.
No.
P-value
Age (years)
188
18 (16-22)
201
20 (17-25)
<0.001
-
NA
-
NA
-
Sex (male)
-
NA
-
NA
-
299
145 (48.5)
90
56 (62.2)
0.022
Sore throat
186
132 (71.0)
201
145 (72.1)
0.798
298
242 (81.2)
89
35 (39.3)
<0.001
Malaise
187
134 (71.7)
201
150 (74.6)
0.509
298
215 (72.1)
90
69 (76.7)
0.396
Abdominal pain
187
58 (31.0)
201
50 (24.9)
0.178
298
77 (25.8)
90
31 (34.4)
0.110
Nausea or vomiting
187
56 (29.9)
200
42 (21.0)
0.043
297
79 (26.6)
90
19 (21.1)
0.294
Lymphadenopathy
187
142 (75.9)
200
147 (73.5)
0.582
297
246 (82.8)
90
43 (47.8)
<0.001
Previous corticosteroid therapy
185
15 (8.1)
200
10 (5.0)
0.216
295
22 (7.5)
90
3 (3.3)
0.252
Serum bilirubin (mg/dL)
188
0.70 (0.50-1.40)
201
0.90 (0.56-1.82)
0.016
299
0.80 (0.50-1.50)
90
0.90 (0.50-1.90)
0.247
Serum AST (IU/L)
187
130 (57-213)
201
99 (49-191)
0.174
298
114 (55-191)
90
104 (49-263)
0.778
Serum AST (times the ULN)
187
4.64 (2.16-7.69)
201
3.68 (1.80-7.08)
0.125
298
4.04 (1.92-7.12)
90
4.02 (1.76-9.88)
0.637
Serum ALT (IU/L)
187
232 (100-360)
201
210 (92-356)
0.541
298
224 (107-360)
90
212 (87-386)
0.690
Serum ALT (times the ULN)
187
7.72 (3.20-11.0)
201
6.41 (2.96-10.6)
0.376
298
6.70 (3.06-10.6)
90
6.81 (2.72-11.3)
0.965
Serum GGT (IU/L)
187
111 (60-186)
201
120 (54-211)
0.497
298
111 (56-187)
90
136 (54-253)
0.096
Serum GGT (times the ULN)
187
2.92 (1.43-4.90)
201
2.81 (1.28-4.71)
0.499
298
2.81 (1.36-4.44)
90
3.32 (1.20-5.78)
0.184
Serum AP (IU/L)
184
393 (206-542)
197
333 (191-508)
0.143
293
361 (198-529)
88
361 (167-528)
0.380
Serum AP (times the ULN)
184
1.91 (1.02-2.77)
197
1.71 (0.95-2.60)
0.164
293
1.78 (1.00-2.65)
88
1.82 (0.77-2.75)
0.703
Blood leukocytes (x109/L)
188
10.4 (6.9-14.7)
201
12.9 (8.0-17.2)
0.004
299
12.5 (8.16-16.3)
90
9.83 (5.95-14.3)
0.003
Blood lymphocytes (%)
188
53.0 (42.7-61.3)
201
51.8 (40.8-61.7)
0.679
299
53.3 (43.6-61.5)
90
45.8 (25.0-60.0)
0.006
Serum IgG (mg/dL)
127
1350 (1125-1600)
130
1330 (1060-1580)
0.384
189
1370 (1140-1610)
68
1155 (978-1465)
0.001
Serum IgA (mg/dL)
127
282 (195-354)
130
275 (187-363)
0.852
189
282 (192-363)
68
268 (207-355)
0.651
Serum IgM (mg/dL)
127
252 (202-355)
130
214 (154-309)
0.003
189
245 (188-338)
68
189 (130-274)
0.001
Serum β2-microglobulin (mg/L)
52
4.31 (3.39-5.70)
51
4.76 (3.95-6.21)
0.152
74
4.30 (3.53-5.64)
29
5.50 (4.24-7.02)
0.008
Data are medians and interquartile ranges (within parentheses) or absolute numbers and percentages (within parentheses). No., number of patients with available determination. P-values were obtained with the Mann-Whitney test or the chi-square test. AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; AP, alkaline phosphatase; ULN, upper limit of normal. NA, not applicable.
The patients with increased serum bilirubin concentrations showed higher levels of immune activation markers than the patients with normal bilirubin concentrations. Blood lymphocyte counts, serum IgM concentrations, and serum β2-microglobulin concentrations were higher in the patients with increased serum bilirubin levels (Table 1). Immune activation markers were also correlated with liver injury markers (Table 3). Blood leukocyte counts, lymphocyte counts, and serum IgM concentrations were correlated with serum AST, ALT, GGT, and AP. Markers of cholestasis (serum GGT and AP) were significantly correlated with serum β2-microglobulin concentrations. Furthermore, serum AP was also correlated with serum IgG and IgA (Table 3).
Table 3Correlation between liver injury markers and immune markers in patients with infectious mononucleosis.
Serum bilirubin concentrations were not significantly associated with the positivity of heterophile antibody test, which was available for 369 patients (Table 4). The heterophile-positive patients were younger than the heterophile-negative patients, and more frequently presented sore throat and lymphadenopathy. The heterophile-positive patients also showed higher blood lymphocyte counts, higher serum transaminase levels, higher serum AP levels, and higher concentrations of serum immunoglobulins, particularly IgM (Table 4).
Table 4Demographic, clinical, and biological characteristics of patients with infectious mononucleosis, stratified by the results of the heterophile antibody test.
Heterophile antibody test
Positive
Negative
P-value
No.
No.
Age (years)
327
18 (17-22)
42
25 (19-34)
<0.001
Sex (male)
327
168 (51.4)
42
22 (52.4)
0.902
Sore throat
326
257 (78.8)
42
13 (31.0)
<0.001
Malaise
327
235 (71.9)
42
33 (78.6)
0.359
Abdominal pain
327
89 (27.2)
42
12(28.6)
0.853
Nausea or vomiting
326
80 (24.5)
42
13 (31.0)
0.368
Lymphadenopathy
326
266 (81.6)
42
20 (47.6)
<0.001
Previous corticosteroid therapy
325
20 (6.2)
41
3 (7.3)
0.772
Serum bilirubin (mg/dL)
327
0.8 (0.5-1.7)
42
1.0 (0.5-1.5)
0.559
Serum AST (IU/L)
326
123 (58-204)
42
81 (40-232)
0.069
Serum AST (times the ULN)
326
4.42 (2.16-7.52)
42
3.15 (1.45-6.27)
0.071
Serum ALT (IU/L)
326
235 (126-364)
42
129 (43-223)
<0.001
Serum ALT (times the ULN)
326
7.44 (3.68-11.3)
42
4.18 (1.48-7.68)
<0.001
Serum GGT (IU/L)
326
116 (61-203)
42
91 (34-181)
0.071
Serum GGT (times the ULN)
326
2.97 (1.55-4.89)
42
1.87 (0.89-4.42)
0.042
Serum AP (IU/L)
320
371 (205-556)
41
279 (124-428)
0.007
Serum AP (times the ULN)
320
1.89 (1.02-2.74)
41
1.29 (0.61-2.13)
0.017
Blood leukocytes (x109/L)
327
12.7 (8.6-16.7)
42
6.1 (4.2-10.8)
<0.001
Blood lymphocytes (%)
327
54.0 (44.2-62.0)
42
44.0 (25.7-54.6)
<0.001
Serum IgG (mg/dL)
211
1370 (1135-1610)
31
1150 (944-1555)
0.014
Serum IgA (mg/dL)
211
295 (202-373)
31
230 (155-302)
0.009
Serum IgM (mg/dL)
211
243 (188-338)
31
170 (142-252)
<0.001
Serum β2-microglobulin (mg/L)
86
4.49 (3.85-6.13)
12
4.78 (4.07-6.05)
0.862
Data are medians and interquartile ranges (within parentheses) or absolute numbers and percentages (within parentheses). No., number of patients with available determination. P-values were obtained with the Mann-Whitney test or the chi-square test. AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; AP, alkaline phosphatase; ULN, upper limit of normal.
Three patients died during their hospitalization, developing sepsis-related multiorgan failure in these three cases in the context of (1) a 17-year-old male patient with diabetes mellitus and severe previous disability (baseline serum bilirubin, 0.3 mg/dL); (2) a 33-year-old female patient with an acute natural killer/T-cell lymphoproliferative disorder that developed simultaneously with the detection of EBV infection (baseline serum bilirubin, 3.7 mg/dL); and (3) a 47-year-old male with EBV-related hemophagocytic syndrome (baseline serum bilirubin of 1.1 mg/dL that reached 7.4 mg/dL during hospitalization). The hospital stay was longer for the patients with increased serum bilirubin concentrations (Table 1).
The liver abnormalities normalized quickly in all patients. Among the patients with serum bilirubin concentrations ≥3 mg/dL on admission, a second bilirubin determination was available for 45 cases in the following two months of follow-up (median, 27 days; range, 14-67 days). Serum bilirubin concentrations significantly decreased in all cases and had already normalized (<1.5 mg/dL) in 35 of the patients (77.7%). The remaining 10 patients (corresponding to the patients with the highest baseline bilirubin concentrations and one patient with Gilbert's syndrome) maintained minor bilirubin elevations that subsided in the following weeks. On average, the median rate of serum bilirubin decrease in these 45 cases was 15.0% per day (IQR, 10.5%-23.7% per day). Serum transaminase alterations tended to normalize in parallel with serum bilirubin normalization (data not shown).
Discussion
The selected case presented a low-grade lymphoproliferative disorder (monoclonal B lymphocytosis of CLL-type with IgM-lambda gammopathy)
and acute EBV infection, which was confirmed both serologically and by demonstrating of blood EBV-DNA. Lymphocytosis was present before EBV infection and monoclonal gammopathy was detected from the time of admission for infectious mononucleosis, suggesting that monoclonal B lymphocytosis existed prior to EBV infection. Patients with similar lymphoproliferative disorders are not particularly prone to primary EBV infection,
probably because EBV contact and subsequent mononucleosis tend to occur in children or young adults, while CLL-type disorders tend to occur in older individuals. The patient had neither a previous nor subsequent history of infections indicative of immune deficiency, but developed a severe EBV-induced hepatitis, which resembled that observed in immunosuppressed transplant recipients
The liver biopsy confirmed the characteristics of EBV-induced hepatitis, including portal infiltration, centrilobular necrosis, and, particularly, sinusoidal infiltration by T cells in a beadlike or “Indian filing” pattern.
Interestingly, T cells infiltrating the liver (but not hepatocytes) are infected by EBV in cases of severe hepatitis, thus suggesting that liver damage might be immune-mediated.
In the previous and subsequent asymptomatic periods, the presented patient experienced blood lymphocytosis and elevated serum concentrations of IgM and β2-microglobulin that increased by 3-fold, 5-fold, and 3-fold, respectively, during the acute mononucleosis hepatitis. The reported case also had increased serum triglycerides during the acute phase, a marker of activation in immunoproliferative disorders
Consistent with this notion, elevated serum bilirubin concentrations were associated with these immune activation markers in the entire sample of 389 adult patients with EBV-induced infectious mononucleosis, an acute self-limiting lymphoproliferative disorder.
Infectious mononucleosis is thought to be immunopathological in nature, because of an exaggerated T cell response to antigens expressed on infected B lymphocytes. Previous studies observed that the variability in clinical features of infectious mononucleosis directly relates to the level of T cell activation,
The proportion of patients with infectious mononucleosis and hyperbilirubinemia in this series (serum bilirubin ≥1.5 mg/dL in 27.2% of cases, and >3 mg/dL [the standard threshold for clinical jaundice] in 12.0% of cases) was higher than that previously reported.
The patients in this series were admitted to the hospital, thus selecting the most severe cases. Moreover, all patients were older than 15 years. It is well known that primary EBV infection is more often clinical as infectious mononucleosis in adulthood than in childhood,
Our reported case may be an example, although no clear association between age and hyperbilirubinemia was observed in the entire sample of adults in the present series. The heterophile antibody-negative patients were older than the positive patients. Serum bilirubin concentrations were similar between the heterophile-positive and negative individuals; however, the heterophile-positive patients showed higher levels of liver damage markers, blood lymphocytes, and serum immunoglobulins, further linking liver disease with immune response in patients with infectious mononucleosis.
The clinical presentation of the reported case is characteristic in that the typical symptoms of infectious mononucleosis (such as tonsillitis and subsequent sore throat) were absent, whereas digestive symptoms were prominent. In the entire series, digestive symptoms (nausea and vomiting) and abdominal pain were positively associated with hyperbilirubinemia while sore throat was negatively associated. These findings suggest that patients with EBV-induced hepatitis show a distinct clinical syndrome, as previously suggested by smaller samples of cases.
In the reported case, corticosteroids were administered along with acyclovir for severe acute hepatitis. The administration of corticosteroids prior to admission was associated with a lower degree of hyperbilirubinemia, but this might be due to a selection bias because the most frequent indication for corticosteroids is tonsillitis hindering swallowing, and sore throat was negatively associated with hyperbilirubinemia. With the exception of the two patients who died from EBV-related lymphoproliferative disorders, the resolution of the hepatitis and hyperbilirubinemia was rapid and satisfactory in all cases, as described earlier.
Among the patients with bilirubin concentrations >3 mg/dL, the rate of decrease in serum bilirubin concentrations was similar in those who took corticosteroids during admission and those who did not (data not shown).
The present study has the strengths of sample size and follow-up (retrospective cohort design). As weaknesses, the study is observational and retrospective.
All of the patients were adults (older than 15 years) and were admitted to the hospital. The findings can therefore only be applied to patients with infectious mononucleosis of similar ages and severity. Finally, the analysis of immune activation in our series was limited to the quantification of immunoglobulins, β2-microglobulin, and lymphocyte count, but more precise markers of lymphocyte activation were not available.
In summary, transient hyperbilirubinemia is common during infectious mononucleosis in hospitalized adult patients. Patients with hyperbilirubinemia have a distinct clinical presentation with less frequent pharyngitis symptoms and more frequent digestive symptoms and abdominal pain. Hyperbilirubinemia during infectious mononucleosis was associated with immune activation markers. Further studies are needed to determine in detail the mechanisms of hepatic damage during primary EBV infection.
Authors contributions
A.G.Q. and J.C.F. conceived and designed the analysis; E.P.G, R.A., and C.B. collected the data; H.L. performed histological studies and contributed to laboratory data interpretation; J.C.F and A.G.Q performed statistical analysis; E.P.G. and A.G.Q wrote the paper. All authors revised the final version of the manuscript.
Source of funding
The study had no specific funding.
Declaration of Competing Interest
The authors have no conflict of interest to declare.
00488-8&id=gr1.jpg){kind=link}
00488-8&id=gr2.jpg){kind=link}