Serum amyloid A protein in the course of infectious mononucleosis.

Background: Serum amyloid A (SAA) protein is a major acute phase protein. Increased concentrations have been reported in many inflammatory diseases. In bacterial infections, high levels correlate with those of C-reactive protein (CRP). In viral infections, where CRP changes are weaker, SAA is of value for establishing early diagnosis, monitoring the severity, and the evolution of the disease. Objective: Evaluation of SAA as a marker for diagnosis of infectious mononucleosis, including severe forms. Material and methods: A total of 31 patients with non-complicated and severe, complicated infectious mononucleosis were examined. SAA and CRP were measured by immuniturbidimetric assays at the day of admission and 4.97 ± 1.35 days later.
Results: SAA increases significantly than those in a control group, without correlation with the etiologic agent. It decreases when full recovery appears. In the subgroup of subjects with complications, we observed significant increased SAA when Epstein-Barr virus /EBV/ was the etiologic agent, in the course of bacterial and viral secondary infection. SAA is higher than CRP in non-complicated group. In cases of bacterial superinfections, both increase simultaneously and treatment have to be adapted. Second, serum sample for CRP is normal in patients without full recovery where SAA stay increased.
Conclusion: In viral infections, high SAA concentrations are indicative for early diagnosis, severe course of the diseases, effect of the treatment, early recovery, and disease outcome. When SAA and CRP increase simultaneously, bacterial co-infection is suspected, and relevant antibiotic treatments have to be initiated.

Introduction

Serum amyloid A (SAA) protein is a major acute phase protein (APP), which belongs to the apolipoprotein family.[1] During the acute phase response (APR), many cytokines induce its production by hepatocytes mainly,[2] but extrahepatic production has also been documented.[3] For many years, SAA was a protein without function, but today is known that, in the course of acute inflammation, it stimulates the immune cell chemotaxis[4] and cytokines production, inhibits bacterial adhesion to epithelial cells,[4] and has a direct antimicrobial effect.[5] The clinical significance of SAA is analyzed by many authors,[2,6,7] but its role in laboratory diagnosis of infectious diseases is still poorly evaluated. As a positive APP, SAA concentrations in acutely infected patients increase rapidly up to 1000-fold in the first 3–6 h, peaking on the third day and returning to the baseline on day four.[8] In bacterial infections, SAA reaches a sera concentration about 2000 mg/l, in parallel with increasing levels of C-reactive protein (CRP), but more promptly than it,[1,9] correlating with severity.[10] In viral infections, CRP changes are in a narrower range, so a more sensitive marker of inflammation is needed for establishing early diagnosis, severe clinical course, and presented complications.

Aim

In this study, the authors evaluate SAA as a potential laboratory marker for diagnosis of viral and severe viral infections. To achieve it, we based on the next points: (1) changes in the sera concentration of SAA in the course of non-complicated infectious mononucleosis; (2) changes in the sera concentration of SAA in the course of severe, complicated infectious mononucleosis; and (3) comparison between SAA and CRP as mentioned above.

Material and methods

Patients

The study group consists 31 patients (n = 31), diagnosed as infectious mononucleosis and hospitalized in the Department of Infectious Diseases, University Hospital “St. Marina,” Varna, Bulgaria, during December 2014–February 2016. On the base of the causative agent, two subgroups are divided: Epstein–Barr virus (EBV)-caused mononucleosis (n = 16) and cytomegalovirus (CMV)-caused mononucleosis (n = 15). Both the subgroups are separated to non-complicated cases (EBV, n = 10; CMV, n = 9) and severe, complicated cases (EBV, n = 6; CMV, n = 6).

All of the subjects were between 1 and 30 years of age (Table 1), without suffering chronic diseases, and with no anamnesis for operation procedures or vaccination in the past 6 months. The basis of including patient from that age is associated with the fact that in the developed world, primary infection with EBV and CMV appears during child and adulthood mainly, resulting in symptomatic infectious mononucleosis. As a contrast, in developing world, because of the poor sanitation control, primary infection is realized early, up to 1 years of age, and it is commonly asymptomatic.[11]

Table 1.

Mean age of the subjects.

Etiologic agent	EBV, n = 16		CMV, n = 15
Form	Non-complicated, n = 10	Complicated, n = 6	Non-complicated, n = 9	Complicated, n = 6
Mean age ± SD	6.1 ± 4.32	8.83 ± 11.29	3.67 ± 2.36	8.17 ± 5.30

CMV, cytomegalovirus; EBV, Epstein–Barr virus; SD, standard deviation.

Determining the severe form of infectious diseases, we used the definition of Radev and colleagues, 1983—“the severity of the disease depends on the grade of intoxication and organ complication, including circulatory shock.”

All of our patients were treated using intravenous glucose solutions, vitamins, antihistamines, and antipyretics. When bacterial infections were suspected, empiric antibiotic treatment was started, using wide-spectrum antibacterial drugs. In the course of severe infectious mononucleosis, steroid, commonly methilprednisolon 1 mg/kg/24 h in tapered over 5–6 days was added.

Methods

Two serum samples were obtained from all of the subjects—on the day of admission, indicating the acute phase of the disease, and 4.97 ± 1.35 days later, in parallel with the beginning of convalescent stage. A small volume of (5 ml) venous blood was collected by Vacutainer system from all the subjects, for determining SAA and CRP. Serum was separated after centrifugation in 3500g for 10 min, using HERAEUS, Kendro, Germany centrifuge. Sera were stored at −20°C until analysis. Single serum sample was also obtained from the healthy controls.

SAA and CRP were measured by immuniturbidimetric assays, adapted on Olympus AU400, AU400 Chemistry Analyzer. Reference ranges for SAA are up to 10 mg/l and for CRP—up to 5 mg/l, according to the mentioned techniques. Etiologic diagnosis was proved by clinic and epidemiologic observation and serological analysis for anti–viral capsid antigen (VCA) immunoglobulins M and G (IgM/IgG) against EBV and anti-IgM/IgG against CMV on the day of admission.

All the laboratory analyses were done in the relevant laboratory of “St. Marina” Hospital, Varna, Bulgaria.

Results

Mean age of the subjects and sex distribution are shown on Table 1 and Figure 1.

Figure 1.

Sex distribution of the subjects.

Control group consists of healthy subjects (n = 30), aged between 1 and 30 years, 13 males and 17 females (Table 2).

Table 2.

Control group.

Subjects	Male	Female	Mean age, SD	Mean SAA, SD	Mean CRP, SD
n = 30	13 (43.33%)	17 (56.67%)	20.17 ± 9.12	3.08 ± 1.25 mg/l	4.21 ± 2.21 mg/l

CRP, C-reactive protein; SAA, serum amyloid A; SD, standard deviation.

1. Changes in the sera concentration of SAA in the course of non-complicated EBV and CMV cases

On the day of admission, mean value of SAA was 214.72 ± 148.02 mg/l in the group of non-complicated EBV versus 288.77 ± 262.91 mg/l in the group of non-complicated CMV. With regard to the second sample, it was in a referent range in 6 (60%) of EBV and 6 (66.67%) of CMV patients. SAA was abnormal in all of the subjects from the both group, who have been not fully recovered and have had a fever and angina at the time of the second serum sample collection (Table 3).

Table 3.

SAA and CRP in studied groups.

Agent	EBV, n = 16				CMV, n = 15
Group	Non-complicated, n = 10		Complicated, n = 6		Non-complicated, n = 9		Complicated, n = 6
Phase	Acute	Convalescence	Acute	Convalescence	Acute	Convalescence	Acute	Convalescence
SAA, mg/l ± SD	214.72 ± 148.02	17.86 ± 18.19	292.1 ± 193.57	31.75 ± 36.50	288.77 ± 262.91	19.35 ± 38.50	256.1 ± 273.05	18.52 ± 16.10
CRP, mg/l ± SD	51.04 ± 53.96	4.23 ± 3.89	55.75 ± 50.59	10.56 ± 16.71	39.57 ± 33.78	3.14 ± 3.55	51.84 ± 72.00	3.66 ± 3.27

CMV, cytomegalovirus; CRP, C-reactive protein; EBV, Epstein–Barr virus; SAA, serum amyloid A; SD, standard deviation.

Mean value of SAA in the control group was 3.08 ± 1.25 mg/l.

2. Changes in the sera concentration of SAA in the course of severe, complicated EBV and CMV cases

In the EBV-caused mononucleosis subgroup, we registered six patients with complications: 2 (33.33%) with bronchitis, 2 (33.33%) with rotavirus enteritis, 1 (16.67%) with furunculosis, and 1 (16.67%) with gingivitis. The mean value of SAA from first and second sample are shown in Table 3. Its highest concentration was 533.7 and 525.5 mg/l in children who were co-infected with rotavirus (so-called patient 1 and patient 2, respectively). In those cases, second measurement of SAA (during convalescence) shows value of 103.8 and 12.5 mg/l, respectively. On the day of obtaining the second serum sample, patient 1 had no sign and symptoms of mononucleosis, but acute enteritis and low grade hypokalemia had not been resolved.

In the subgroup of CMV-caused mononucleosis, complications such as liver involvement (hepatitis) was found in 4 (66.66%), herpes simplex labialis in 1 (16.67%), and Staphylococcus aureus–caused acute tonsillitis in 1 (16.67%). SAA values were 90.2, 49.8, 111.9, 6.8, 614.7, and 663.2 mg/l versus 10.7, 16.3, 17.8, 1.7, 52.6, and 12.0 mg/l, respectively, in the second serum samples.

3. Comparison between SAA and CRP in studied groups

When SAA and CRP were compared, some important findings were established (Table 3). In EBV- and CMV-caused non-complicated cases, mean value of CRP was 23.78% and 13.70% lower than SAA. Thus SAA/CRP ratio was higher than those in healthy control group (mean value of CRP in the control group was 4.21 ± 2.21). In second serum samples, CRP was normal in seven subjects with EBV mononucleosis and in eight with CMV. Daily evaluation of these patients have shown three EBV, and three CMV cases still had different complaints, and its SAA values were up to the referent ranges, correlating with lack of full recovery.

In regards to the severe, complicated cases, CRP changes followed those of SAA, but in a narrower range—SAA/CRP ratio was high but lower than those in non-complicated cases. Maximum CRP value was 204.72 mg/l in a CMV-infected patient with Staphylococcal tonsillitis, while SAA was three times greater (663.2 mg/l). Analyses of the second sample have shown CRP in a referent range, but SAA more than 10 mg/l, in accordance with persistent sore throat and the signs of follicular angina.

Discussion

Infectious mononucleosis is an acute infectious disease, with EBV as etiologic agent in 90% of cases. CMV, another herpesvirus, is commonly associated with EBV-negative infectious mononucleosis syndrome.[12] The disease takes its clinical course with angina, enlargement of lymph nodes, liver and spleen, and specific laboratory constellation.[13] No matter if it is viral-caused disease, CRP changes may observe, precipitating the exact differential diagnosis.[14] Because of this, a more specific laboratory marker is necessary to be used in the practice, early, and promptly designating inflammation from viral origin.

Our study demonstrates that in the beginning of non-complicated infectious mononucleosis, SAA concentration increases than those in a control group, reaching a mean value of 214.72 ± 148.02 in EBV and 288.77 ± 262.91 in CMV cases. Similar changes have been observed in another case study, where mean SAA levels in patients with infectious mononucleosis were 363.4 ± 287.8.[6]

According to Miwata and colleagues,[15] decreasing of SAA depends on the recovery processes and may be used as a prognostic marker for disease outcome. Our results show normal SAA in six EBV and six CMV patients, who were without any complaints at the day of second serum sample collection versus SAA above referent ranges in the rest, who still had been ill at that time.

SAA may be used as a potential marker for different types of complications, commonly secondary infections. Higher levels of SAA were registered in the course of bronchitis, furunculosis, Staphylococcal tonsillitis, and when co-infection with other viruses appears, suggesting mobilization of different diagnostic tools and implication of wide-spectrum treatment. So we think, measurement of SAA is of value in monitoring severe, complicated cases of infectious mononucleosis. We expected SAA concentration to be significantly higher in those with liver involvement, basing to the main hepatic production of this APP.[16] But our results indicate a mean SAA value of 64.67 ± 40.16 mg/l which were lower than those in other mentioned complication.

Treatment efficiency may be checked by decreased SAA levels, established 4 days later.[5] In our study, SAA concentration from the second serum sample, obtained 4.97 ± 1.35 days later, were normal in four patients (two with complicated EBV and two with complicated CMV), who were fully recovered at that time. Therefore, persistent elevations of SAA beyond the 4–6 days could be related to severe infections, infectious complications, or ineffective treatment. Thus, the SAA could be used as a specific marker which reveals the need of treatment changing.

According to some authors, CRP is a diagnostic marker for bacterial infections, and its serum changes follow those of SAA.[14] We also compared SAA and CRP changes and SAA/CRP ratio in non-complicated and complicated EBV- and CMV-caused mononucleosis. Within the former group, SAA was higher than CRP, and SAA/CRP ratio was high, indicating that SAA is a more sensitive marker for viral inflammation. This fact was previously suggested by Shainkin-Kestenbaum and colleagues.[17] In complicated EBV and CMV groups, maximum CRP levels were observed in patients with furunculosis and Staphylococcal tonsillitis (up to 204.72 mg/l), while its concentrations varies between 0.9 and 69.11 mg/l. Therefore, when SAA and CRP concentrations both significantly increases in the course of viral disease, a mixed infection with bacterial agent has to be suspect, and antibiotic treatment have to be initiated. CRP was normal in 12 of 16 EBV cases and 13 of 15 CMV cases, no matter that seven patients in each group still had complaints, and because of this, it could not be used as an early marker of recovery process, unlike SAA.

Conclusion

In viral infections, high SAA concentrations are of value for early diagnosis, severe course of the diseases, effect of the treatment, early recovery, and disease outcome. When SAA and CRP raise simultaneously, bacterial co-infection is suspected, and relevant antibiotic treatment have to be initiated.