Received: October 11, 2016 | Accepted: January 11, 2017 | Published: January 21, 2017
OBM Hepatology and Gastroenterology 2017, Volume 1, Issue 1, doi:10.21926/obm.hg.1701001
Academic Editor: Ikuo Nakamura
Recommended citation: Kamezaki H, Kanda T, Yokosuka O, Tsukahara M, Kasahara Y, Sakurai T. Pneumocystis jirovecii pneumonia Suspected Due to Low-Dosage Prednisolone Treatment for Acute Autoimmune Hepatitis: Case Report and Literature Review. OBM Hepatol Gastroenterol 2017;1(1):001; doi:10.21926/obm.hg.1701001.
© 2017 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.
The Japan Ministry of Health, Labor, and Welfare have published a Diagnostic Guide for Autoimmune Hepatitis (AIH) written in Japanese and edited by the group focused on “Studies Concerning Intractable Hepatic and Biliary Disease”. This advocates the administration of sulfamethoxazole/trimethoprim (ST) combination therapy in the form of high doses of steroid drugs to prevent Pneumocystis jirovecii pneumonia (PCP). However, clearly defined guidelines for the initiation and termination of therapy are not provided. ST combination drugs are associated with side-effects such as liver damage; therefore, it is necessary to consider the suitability of this treatment for each individual patient. For patients not infected by HIV, a prednisolone (PSL) dose of ≥20 mg/day has been recommended as the standard for the initiation of PCP preventative therapy, but there is no clear global consensus on the standards for AIH patients. Here, we present a case report that may aid in the determination of criteria for preventative ST combination therapy in these patients.
The patient in this case was a 71-year-old woman. She had originally consulted a local clinician with chief complaints of dark-colored urine, pale stools, and decreased appetite with a duration of these symptoms of 10 days. The patient had a medical history of hypertension and had been prescribed telmisartan for several years but had never been advised to undergo detailed examination following medical check-ups. She had not received any blood transfusions and did not report any allergies. There was no family history of hepatic or autoimmune diseases. The patient had no history of tobacco use and consumed alcohol only on a social basis. She had been drinking Jukusei yazuya vinegar® (Yazuya Inc., Fukuoka City) for several years. Blood test results revealed elevated hepatobiliary enzymes, and the patient was referred to our department and admitted to the hospital the following day for further tests and treatment.
At the time of admission, the patient had the following clinical characteristics: height, 148.4 cm; weight, 66.3 kg; blood pressure, 119/58 mmHg; pulse rate, 71 beats per minute; heart rhythm, regular; temperature, 36.9°C; and room-air SpO2, 98%. Physical examination revealed jaundice of the skin and bulbar conjunctiva, with no evidence of limb edema or spider angioma. Normal breathing sounds were auscultated. The abdomen was not distended; in addition, it was soft to the touch and the liver and spleen were not palpable.
Blood tests at the time of admission (Table 1) revealed elevated hepatobiliary enzyme levels (total bilirubin [T-Bil], 18.0 mg/dL; conjugated bilirubin [C-Bil], 13.8 mg/dL; aspartate aminotransferase [AST], 3,111 U/L; and alanine aminotransferase [ALT], 1,586 U/L), as well as increased prothrombin time (PT; 52%) and increased ammonia level (NH3; 73 μg/dL), suggesting reduced hepatic function. The patient tested negative for hepatitis A, B, C, and E viruses, and neither Epstein–Barr virus nor cytomegalovirus was indicated. The patient was also HIV antibody-negative. Immunoglobulin (Ig) fractions showed increased levels of IgG (2,927 mg/dL) and IgM (334 mg/dL). Autoimmune antibody tests were positive for antinuclear antibody (ANA) at a dilution of 1:160 (homogeneous and speckled staining patterns), but were negative for anti-smooth muscle antibody, anti-LKM-1 antibody, anti-mitochondrial M2 antibody, and anti-ds-DNA IgG. Drug-induced lymphocyte stimulation tests (DLSTs) were negative for telmisartan and Jukusei yazuya vinegar®.
Table 1 Laboratory data.
Abbreviations: WBC, white blood cell; Neut, neutrophil; Lymph, lymphocyte; Eosino, eosinophil; Hb, hemoglobin; Plt, platelet; TP, total protein; Alb, albumin; T-Bil, total bilirubin; C-Bil, conjugated bilirubin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH, lactate dehydrogenase; ALP, alkaline phosphatase; γ-GTP, gamma-glutamyl transpeptidase; BUN, blood urea nitrogen; Cre, creatinine; UA, uric acid; NH3, ammonia; FPG, fasting plasma glucose; HbA1c, glycosylated hemoglobin; AFP, alpha-fetoprotein; PIVKA-II, protein induced by vitamin K absence-II; PT, prothrombin time; CRP, C-reactive protein; IgG, immunoglobulin G; IgM, immunoglobulin M; IgM-HA Ab, immunoglobulin M-hepatitis A antibody; HBs Ag, hepatitis B surface antigen; HBs Ab, hepatitis B surface antibody; HBc Ab, hepatitis B core antibody; IgM-HBc Ab, immunoglobulin M-hepatitis B core antibody; HBV-DNA, hepatitis B virus-deoxyribonucleic acid; HCV Ab, hepatitis C virus antibody; HCV-RNA, hepatitis C virus-ribonucleic acid; IgA-HEV Ab, immunoglobulin A-hepatitis E virus antibody; EBV-VCA-IgM, Epstein–Barr virus-viral capsid antigen-immunoglobulin M; CMV-IgM, cytomegalovirus-immunoglobulin M; HIV Ag/Ab, human immunodeficiency virus antigen/antibody; ANA, antinuclear antibody; ASMA, anti-smooth muscle antibody; LKM-1 Ab, liver-kidney microsome-1 antibody; AMA-M2, anti-mitochondrial-M2 antibody; C3, complement 3; C4, complement 4; CH50, 50% hemolytic complement activity; ds-DNA IgG, double-stranded-deoxyribonucleic acid immunoglobulin G.
Abdominal computed tomography (CT) revealed thickening of the gallbladder wall and a periportal collar sign (Figure 1).
Figure 1 Abdominal CT. Abdominal CT reveals areas of low density surrounding the portal veins in the liver (periportal collar sign) (A) and subserosal edema of the gallbladder wall (B).
On the basis of the examination and diagnostic test results, acute hepatopathy was diagnosed, with AIH strongly indicated. According to the Simplified Scoring for Autoimmune Hepatitis [(1) ANA positive at 1:160 = 2 points, (2) elevated IgG = 2 points, (3) lack of indication of viral hepatitis = 2 points] the patient scored a total of 6 points . According to the Revised Scoring System of the International Autoimmune Hepatitis Group [(1) female = 2 points, (2) decreased ALP/AST = 2 points, (3) increased IgG = 2 points, (4) ANA positive at 1:160 = 3 points, (5) negative anti-mitochondrial antibody = 0 points, (6) negative for markers of hepatitis virus = 3 points, (7) positive for history of medication use = −4 points, (8) social drinking = 2 points] the patient scored a total of 10 points . Percutaneous liver biopsy was not performed because of the presence of mild ascites on ultrasound examination and prolonged clotting time. In addition, the patient refused transjugular liver biopsy.
Treatment was initiated with 60 mg/day oral prednisolone (PSL) on the day after hospital admission (Day 1) because of concerns caused by the increasing severity of symptoms. Thereafter, aminotransferase and IgG levels decreased gradually, and PSL dosage was reduced by 5 to 10 mg every week or two. Because the patient made good progress, no other immunosuppressant was used in combination with steroids. To bring about a smooth decrease in aminotransferase levels and to increase albumin levels, we added ursodeoxycholic acid and branched-chain amino acids to the treatment regimen. Subsequently, the patient developed a rash, which was thought to be drug-induced, since the rash receded once treatment was stopped. Because of the lengthening duration of PSL administration, ST combination therapy was considered for prevention of PCP; however, administration was withheld because of concerns regarding side-effects including liver damage and drug rash. The PSL dose was reduced to 15 mg/day by Day 55. On Day 60, no respiratory symptoms were observed, the patient was negative for β-D-glucan (<6.000 pg/mL), and imaging revealed no sign of pneumonia. However, on Day 70, the patient reported difficulty in breathing and was found to be positive for β-D-glucan (58.740 pg/mL). Room-air SpO2 was 96%, lactate dehydrogenase was 582 U/L, and sialylated carbohydrate antigen KL-6 was 1,460 U/mL. Chest CT imaging revealed extensive ground-glass opacification in both lungs (Figure 2) and predominantly centrilobular shadows.
Figure 2 Chest CT images. Chest CT reveals ground-glass opacification with a centrilobular distribution.
PCP was suspected, and the patient was started on ST combination therapy. Phlegm was not observed, and the overall status of the patient was relatively stable, with normal oxygen levels maintained solely through the use of supplemental nasal oxygen. On Day 77, bronchoalveolar lavage fluid tested positive for Pneumocystis jirovecii DNA by polymerase chain reaction (PCR) testing, although Pneumocystis jirovecii cells were not identified; a diagnosis of suspected PCP was based on these findings. Significant bacterial growth was not detected in cultures. Blood cytomegalovirus pp65 antigen (C7-HRP) was negative. Subsequently, ST combination therapy was continued, and PCP remission was observed. PSL was gradually reduced to 7.5 mg/day, and a liver biopsy was performed on Day 165. Mild lymphocytic infiltration was observed in the portal region of the liver (Figure 3).
No findings specific to AIH were observed, possibly because the biopsy was performed after the completion of the treatment course. However, post-treatment evaluation using the Revised International Scoring System (remission following treatment = 2 points) yielded a new total of 12 points regarding suspected AIH . At present, remission has been maintained with no signs of recurrence, and the patient continues to receive 2.5 mg/day PSL.
Figure 3 Histological examination. Liver biopsy specimens show mild portal inflammation with lymphocytic infiltration. [Hematoxylin-eosin staining: A (×100); B (×200); C (×1,000)].
Pneumocystis jirovecii is a pathogenic fungus that causes infection in individuals with cellular immunodeficiency. Such immunodeficiencies may be associated with blood disease or malignancy (e.g., HIV infection or leukemia), or with administration of immunosuppressive agents or steroids (e.g., in organ transplantation or connective tissue disease). The pathogen was discovered in 1909 by Chagas  and, as indicated by the term “pneumo” in the genus, the primary site of infection is the lung. Infection in humans was first discovered as interstitial plasma cell pneumonia in children after World War II. Recognition became more widespread in the 1960s in the context of pneumonia in immunocompromized patients, although interest subsided following the subsequent development of safe and effective antibiotics. However, following the emergence of HIV in the 1980s, PCP was again identified as a significant medical and public health concern .
Phair et al.  followed 1,665 HIV patients (who had not yet progressed to AIDS), of which 168 (10.1%) developed symptoms of PCP. Furthermore, the risk ratio of PCP was increased by 4.9 in patients with a CD4+ lymphocyte count of <200/mm3. For this reason, a CD4+ lymphocyte count of <200/mm3 was chosen as a standard for initiating PCP prevention measures in HIV-infected patients.
Treatment guidelines for PCP in non-HIV-infected patients are based on 116 cases reported by Yale et al. . This study was conducted in PCP patients at the Mayo Clinic, with underlying conditions such as blood disorders, organ transplantation, inflammatory diseases, and malignancies. Of these patients, 105 (90.5%) were treated with steroid therapy, which suggested that this regimen had a strong impact on the development of PCP. However, the dose and duration of steroid therapy differed depending on the condition of the patient, indicating that it may not be appropriate to consider all of these conditions to be equivalent with respect to preventative PCP treatment. Furthermore, Godeau et al.  reported a case in which a patient developed PCP secondary to systemic lupus erythematosus (SLE), with no history of steroid or immunosuppressant use and with a normal lymphocyte count. This may have been related to decreased complement levels due to SLE. Cano et al.  described five patients who developed PCP with no underlying conditions. Overgaard et al.  reported that, among 17 non-HIV-infected PCP patients, three (17.6%) exhibited normal CD4+ lymphocyte counts. These cases show that it is important to recognize that the etiology of PCP cannot be fully explained by lymphocyte count or steroid administration alone.
There is a clear need for consensus regarding standards and guidelines for preventative PCP treatment in non-HIV-infected patients. Sepkowitz et al.  reported that preventative PCP treatment should be considered for patients treated with PSL doses of ≥20 mg/day administered for ≥4 weeks, while Fishman  reported that such preventative PCP treatment should be considered for patients treated with PSL doses of ≥20 mg/day administered for a duration of ≥2–3 weeks. These suggestions were based on experience rather than rigorously obtained evidence, but many reviews and editorials have cited 20 mg/day PSL as a benchmark [3,11,12,13]. Arend et al.  reported on the course of steroid treatment, stating that among 72 cases, 31 (43.1%) developed PCP following rapidly reduced steroid dosage. Sepkowitz et al.  also reported that among 113 cases, 79 (69.9%) developed PCP following reduced steroid dosage. One possible reason for this observation is that, in these non-HIV-infected patients, the steroid itself reduced the inflammation associated with PCP (in other words, steroid treatment might have helped to suppress the symptoms of PCP); therefore the reduced steroid dosage was associated with a less marked effect. Further evidence of this is provided by the observation that among the 113 cases, seven (6.2%) developed PCP within 1 month of beginning steroid treatment, with a median maximum PSL dose among all patients of 120 mg/day. This indicates that steroid use affects the incidence of PCP according to (1) current dose, (2) duration of treatment, (3) rapidity of decreases in dosage, and (4) maximum dose.
The question of how best to approach PCP prevention in AIH patients requiring steroid therapy remains. First, it must be considered that unnecessary preventative treatment may lead to drug resistance in Pneumocystis jirovecii . There is also a danger of exacerbating liver damage due to drug side-effects. Furthermore, reports of PCP accompanying AIH are rare, so the risk of PCP in these patients is unclear. Even considering these issues, however, preventative PCP treatment is becoming relatively more common. To determine possible administration standards, PCP data in combination with other autoimmune/inflammatory/connective tissue diseases can be considered for reference. Yale et al.  reported that more than a quarter of 26 PCP patients received ≤20 mg/day PSL, with the lowest dose being 12 mg/day, and Glück et al.  reported that of two of six PCP patients (33.3%) received 15 mg/day PSL. The CD4+ lymphocyte count in all of these patients was <200/mm3. Mansharamani et al.  reported that among eight PCP patients, all had a CD4+ lymphocyte count <250/mm3, with six (75.0%) having a CD4+ lymphocyte count <200/mm3. Godeau et al.  reported total lymphocyte counts <600/mm3 in 10 of 12 cases (83.3%) of Wegener's granulomatosis who developed PCP. These findings indicate that risk factors for PCP include (1) a PSL dose of ≥12–15 mg/day, (2) a CD4+ lymphocyte count of ≤200–250/mm3, and (3) a total lymphocyte count of ≤600/mm3. For patients receiving PSL for extended periods, decreased CD4+ lymphocyte counts may be observed at doses <20 mg/day. Therefore, in cases of AIH, lymphocyte counts, in addition to aminotransferase and IgG levels, should be monitored for PCP prevention. In the present patient, her total lymphocyte count had decreased to 514/mm3 on Day 42 (Figure 4). Following the development of PCP, on Day 99, the patient’s CD4+ lymphocyte count was 175/mm3 at a PSL dose of 15 mg/day. It should be noted that, in this case, the patient developed only minor symptoms of PCP. It may be that severe PCP was prevented by our rapid detection of minor changes and immediate introduction of ST combination therapy. Walzer et al. noted that patients receiving immunosuppressive drugs are typically sick for approximately 1 to 2 weeks before seeking medical attention . Also, because the bacterial load in non-HIV-associated PCP is small, the rate of diagnosis by microscopic examination is low ; this was the situation in the present case and we speculate that a requirement for PCR diagnosis is inevitable in such cases.
Figure 4 Clinical course. (PSL, prednisolone; ALT, alanine aminotransferase; Lymph, lymphocyte; PT, prothrombin time).
For AIH patients on steroid therapy, monitoring of lymphocyte counts should play an important role in the consideration of preventative PCP treatment. Our case report concurs with previously reported findings that long-term steroid treated non-HIV patients with CD4+ lymphocyte counts of ≤200–250/mm3 and total lymphocyte counts of ≤600/mm3 should be considered for PCP prophylaxis.
Medical examination: Hidehiro Kamezaki, Masanori Tsukahara, Yasunori Kasahara.
Drafted manuscript: Hidehiro Kamezaki.
Revised manuscript: Tatsuo Kanda, Osamu Yokosuka, Takayuki Sakurai.
The authors have declared that no competing interests exist.