2025 Volume 266 Issue 4 Pages 381-386
Despite receiving immunosuppressive therapy, a 37-year-old woman with lupus nephritis experienced refractory nephrotic syndrome and severe kidney dysfunction that necessitated peritoneal dialysis. Thereafter, immunosuppressants were tapered because her lupus symptoms improved. Two weeks after prednisolone termination, she was hospitalized because of acute abdominal pain. A physical examination revealed upper abdominal tenderness. Her dialysate was cloudy and comprised an elevated neutrophil count. A coronavirus disease 2019 test yielded positive results. Infectious peritonitis was suspected, coronavirus disease 2019 was diagnosed, and antibiotic and steroid therapies were initiated. The patient’s symptoms improved rapidly, and steroid therapy was discontinued on day 9. On day 10, she experienced abdominal pain and exhibited an elevated monocyte-predominant dialysate cell count despite antibiotic treatment. The dialysate culture results were negative. Lupus peritonitis was suspected because of her slightly elevated dsDNA antibodies and changes in the dialysate cell fraction. Prednisolone was resumed on day 12. Her symptoms improved immediately, and she was discharged on day 30 without complications. This case highlights unique autoimmune activation by coronavirus disease 2019 (COVID-19) that resulted in unexpected pathological shifts and challenging diagnoses. Clinicians should consider various causes of immune flare-ups in patients with COVID-19 and autoimmune conditions.
Systemic lupus erythematosus (SLE) is an autoimmune disease with clinical manifestations that involve multiple organs. Lupus peritonitis (LP) is a rare manifestation of SLE with a lifetime prevalence of 8% to 11% (Prasad et al. 2012). LP typically occurs with flare-ups of diseases with symptoms associated with SLE. Thus, SLE flare-ups with LP as the main manifestation are extremely rare (Prasad et al. 2012).
Coronavirus disease 2019 (COVID-19) is a novel infectious disease that caused a global pandemic and remains a critical health concern. COVID-19 is known to heighten immune responses associated with autoimmune diseases, including SLE (Fallahi et al. 2023; Lim et al. 2023; van Dam et al. 2023). Although some studies have suggested an association between SLE flare-ups and COVID-19 (Schioppo et al. 2022; Naranjo-Millán et al. 2023), such reports regarding patients with SLE on peritoneal dialysis (PD) are rare.
We report the case of a patient with SLE and COVID-19 on PD who presented with sudden abdominal pain and was initially suspected of suffering infectious peritonitis (IP) that possibly progressed to what was considered LP during the clinical disease course.
A 37-year-old woman presented to our hospital with lupus nephritis (LN). She had first presented with menstrual bleeding and skin rashes 10 years prior. Laboratory test results indicated nephrotic syndrome, thrombocytopenia, positive antinuclear antibodies, and positive dsDNA antibodies. Kidney biopsy results indicated LN (International Society of Nephrology/Renal Pathology Society classification: class V), resulting in a diagnosis of SLE. Arthritis, serositis, peritonitis, and pulmonary involvement were not observed. Immunosuppressive therapy was administered, and all symptoms except those of LN improved rapidly. After administration of immunosuppressants including prednisolone (PSL), tacrolimus, mycophenolate mofetil, and rituximab, the clinical SLE symptoms gradually subsided, and active serological responses became undetectable. However, high levels of proteinuria persisted, and LN remission was not achieved, leading to the initiation of PD 11 years after the initial SLE diagnosis. At that time, the patient was administered PSL 10 mg, tacrolimus 2 mg, and mycophenolate mofetil 1,000 mg daily. After the initiation of PD, no SLE symptoms were detected, and the immunosuppressants were tapered and discontinued. Tacrolimus was withdrawn at 6 months, mycophenolate mofetil at 8 months, and PSL at 10 months after the initiation of PD. The patient regularly used diuretic and antihypertensive medications to control fluids during PD. However, despite the intensification in the use of medications and adjustment of the PD regimen, blood pressure and fluid management became increasingly challenging, leading to a gradual worsening of fluid retention. Regarding the cause of progressive hypertension, no hormonal abnormalities indicative of secondary hypertension, including thyroid hormone imbalances, were observed in her past clinical course.
Two weeks after PSL was discontinued, acute abdominal pain and dyspnea developed, and she was admitted to our hospital. The patient had no allergies, and her medical history included only SLE and LN. At that time, she was using furosemide, trichloromethiazide, tolvaptan, spironolactone, amlodipine, and azilsartan. Her continuous ambulatory peritoneal dialysis regimen comprised 1.5-L exchanges four times per day (three bags of 1.5% dextrose and one bag of 7.5% icodextrin dialysate). Her body temperature was 36.9°C, and her blood pressure and heart rate were 234/151 mmHg and 115 beats/min, respectively. Her respiratory rate was 34 breaths/min, and she had a decreased oxygen saturation of 78%. Physical examination results indicated prominent bilateral lung murmurs and upper abdominal pain. Neither dermatitis nor arthritis were observed. The drained peritoneal dialysate appeared cloudy. There were no signs of an exit site or tunnel infection near the PD catheter. The full laboratory test results are presented in Table 1. Her serum C-reactive protein level was mildly elevated, and her brain natriuretic peptide level was significantly increased. Regarding SLE activity markers, the levels of dsDNA antibodies showed a slight increase, and serum C3 levels, while within the normal range, were slightly decreased (Fig. 1). An analysis of the dialysate indicated an elevated, neutrophil-predominant cell count of 18,413/μL (neutrophils, 80%; monocytes, 20%). Chest radiography and computed tomography (CT) examinations showed severe pulmonary edema with little evidence of infectious pneumonia. An abdominal CT examination showed mild intestinal mucosal edema and bowel wall thickening, with no signs of endogenous infection. A COVID-19 antigen test performed at the time of admission yielded positive results. Based on these results, treatment for COVID-19 was initiated with 1,000 mg intravenous methylprednisolone, 6 mg dexamethasone, and remdesivir. Suspecting hypertensive heart failure as the primary cause of respiratory failure, noninvasive positive pressure ventilation was initiated to treat her cardiac failure, and continuous hemodiafiltration was performed to provide renal support and fluid management. Regarding the abdominal pain, since it was severe and neutrophil-predominant, we first suspected PD-related peritonitis and administered antibiotics (intravenous meropenem and intraperitoneal vancomycin).
Her heart failure and peritonitis symptoms improved rapidly with the initial therapy; therefore, steroid treatment was discontinued on day 9 (Fig. 1). However, on day 10, the patient experienced sudden abdominal pain, an elevated C-reactive protein level (to 5.78 mg/dL from a baseline level of 1.7 mg/dL), and an elevated monocyte-predominant dialysate cell count (to 1,651/μL from a baseline level of 163/μL; neutrophils, 11%; monocytes, 89%) despite receiving an ongoing antibiotic regimen. Initial and follow-up dialysate cultures yielded negative results, and another abdominal CT examination did not show any findings of endogenous infection. After considering the slight elevation in the patient’s dsDNA antibody level at the time of admission as well as the changes in her dialysate cell fraction, we suspected that the pathogenesis of her peritonitis might have shifted from infectious to autoimmune. Therefore, steroid treatment was resumed (40 mg/day PSL) on day 12. Her abdominal pain and dialysate turbidity improved immediately, her dialysate cell count decreased from 1,651/µL to 56/μL, and her levels of dsDNA antibodies also decreased rapidly. Thereafter, PSL was tapered, and no further flare-ups were observed. Based on the clinical course, we considered that LP was likely involved in the pathogenesis of the peritonitis. The patient was discharged on day 30 without any major complications.
Laboratory results and lupus serology and dialysate fluid findings.
Clinical course and treatment.
Steroid and antibiotic therapies were initiated for COVID-19 and IP on the day of admission. Peritonitis improved rapidly with the initial therapy. After steroids were discontinued on day 9, the patient experienced sudden abdominal pain as well as an elevated serum CRP level and monocyte-predominant dialysate cell count. Lupus peritonitis was suspected; therefore, PSL 40 mg/day was resumed on day 12. The symptoms and her dsDNA antibody levels decreased rapidly, with no further signs of flare-ups. The patient was discharged on day 30. C3, complement 3; C4, complement 4; CAPD, continuous ambulatory peritoneal dialysis; CHDF, continuous hemodiafiltration; COVID-19, coronavirus disease 2019; CRP, C-reactive protein; dsDNA, double-stranded DNA; mPSL, methylprednisolone; IP, infectious peritonitis; i.v., intravenous; NPPV, noninvasive positive pressure ventilation; PSL, prednisolone; WBC, white blood cell.
This case presents a very rare instance in which LP was suspected as the diagnosis for peritonitis in a PD patient. We consider that abnormal autoimmune activation induced by COVID-19 may have played a role in the development of this rare condition. Recently, numerous reports have highlighted exacerbations of various autoimmune diseases, including SLE, following COVID-19 (Kichloo et al. 2020; Aram et al. 2021; Schioppo et al. 2022). During the clinical diagnosis process, we considered the possibility that the pathogenesis of the peritonitis had shifted from infectious to autoimmune due to the influence of COVID-19-induced autoimmune activation. However, it is challenging to definitively determine the exact pathogenesis in this case, and various hypotheses should be considered.
The first hypothesis is that the initial peritonitis was bacterial, but later transitioned to LP due to autoimmune activation triggered by COVID-19. The initial peritonitis was neutrophil-dominant, severe, and with a turbid dialysate, which is consistent with a typical PD-related IP according to the International Society for Peritoneal Dialysis guidelines (Li et al. 2022). It is plausible to postulate that the antibiotics used in the initial treatment were effective. Regarding the second instance of peritonitis, since it occurred after the discontinuation of PSL and the relapse occurred during the antibiotic course, it is reasonable to assume that the cause was autoimmune rather than infectious. With a history of SLE, LP should indeed be considered as a differential diagnosis for this PSL-reactive peritonitis. The slight fluctuation in SLE activity markers at the time of admission also supported the suspicion of LP development. Additionally, this episode occurred in the context of COVID-19, where the impact of autoimmune activation could not be ignored. Given that this autoimmune activation can occur even in mild cases of COVID-19 (Assar et al. 2022; Petrou et al. 2023; Sumichika et al. 2024), it is crucial to remain vigilant in cases without evident pneumonia, as demonstrated in this patient. While patients with SLE flare-ups generally experience symptoms similar to those observed at the time of their initial diagnosis (Barrera-Vargas et al. 2016), it is somewhat puzzling that our patient flared in the form of LP, which had never been observed during her clinical history. This might be explained by the intestinal inflammation caused by the initial IP acting as the “first hit”. Infections can trigger the onset and flare-ups of SLE (Aslanidis et al. 2008; Deng and Tsokos 2008; Rasmussen et al. 2015), thus supporting one of the suspected causes of the SLE flare-up in our patient. Combined with COVID-19, we hypothesize that a localized intestinal inflammation due to IP acted as the first hit, with increased autoimmune activation due to COVID-19 serving as a second hit, triggering an exacerbated SLE flare centered on the peritoneum. This diagnosis of LP was supported by the alterations in the dialysate cell fraction, which changed from neutrophils to monocytes during the second relapse. Several studies have confirmed that ascites of patients with LP are predominantly composed of non-neutrophil cell types such as lymphocytes or monocytes (Schousboe et al. 1988; Kaklamanis et al. 1991; Ito et al. 2002), which is consistent with the findings of the present case. However, an observation that contradicts this hypothesis is the negative result of the ascitic fluid culture. While culture-negative results in PD bacterial peritonitis occur in 5% to 41% of the cases (Chen et al. 2007), it is somewhat puzzling that no causative bacteria were detected despite the severe peritoneal findings.
The second hypothesis is that the syndrome was LP from the beginning. Multiple immunosuppressive agents were needed at the time of PD initiation to suppress the activity of SLE, and this episode occurred less than a year after the initiation of PD. Additionally, the patient had end-stage kidney disease (ESKD) and selected PD as the modality of renal replacement therapy, which tends to preserve renal blood flow and residual renal function. While the estimated glomerular filtration rate (eGFR) at the start of PD was relatively low (3 mL/min/1.73 m²), the decline in eGFR was gradual since the initiation of PD. Unfortunately, the preservation of residual renal function might have counteracted the burnout of SLE activity typically observed in patients with ESKD (Coplon et al. 1983; Nossent et al. 1990). Thus, it is plausible that the baseline SLE activity was not entirely suppressed. Furthermore, this episode occurred after discontinuation of all immunosuppressive drugs, a situation in which SLE disease activity is more likely to increase. The slight fluctuations in SLE activity markers at the time of admission also support this interpretation. Taking these aspects into account, the improvement in the initial peritonitis might have been due more to the steroid treatment initiated after admission than to the antibiotics. Thus, the underlying SLE itself might have been exacerbated, leading to the development of LP. However, there are a few observations that do not quite align with this interpretation. One is that the initial peritonitis was quite severe. Considering the stable course of SLE with a gradual tapering of immunosuppressants, the severity and the rapid occurrence of LP appeared unusually intense for a typical SLE flare. Moreover, this hypothesis cannot clearly explain the mechanism by which a baseline SLE flare manifested as LP. The fact that the cell count from the ascitic fluid at the time of admission was predominantly neutrophilic is also somewhat uncommon. Of course, given that the trigger of abnormal immune activation was possibly the infection with SARS-CoV-2, it would not be surprising if such atypical clinical findings were observed.
The third hypothesis is that the initial peritonitis was caused by COVID-19 enteritis and later transitioned into LP. COVID-19 is known to exhibit gastrointestinal symptoms in addition to fever and respiratory symptoms (Yao et al. 2022). The most commonly observed CT finding is bowel wall thickening, observed in approximately 10% of the cases (Onur et al. 2023), which is consistent with the CT findings in this case. The improvement following treatment for COVID-19 also supports this interpretation. Therefore, gastrointestinal inflammation caused by COVID-19 may have been present at the time of admission, with COVID-19 enteritis itself potentially serving as the trigger for the initial peritonitis, eventually leading to the development and exacerbation of LP through autoimmune activation. However, little is known about the association between COVID-19 enteritis and LP development, and it remains uncertain whether COVID-19 enteritis could cause such severe peritonitis. Further accumulation and analysis of cases will be necessary in the future.
As mentioned above, there are various possible interpretations for this pathological condition, and it is difficult to definitively determine the exact progression of the disease. However, given that this pathological progression occurred in the presence of COVID-19, it is plausible that the disease might have played a significant role in the onset and exacerbation of LP, possibly via autoimmune activation.
Our study has a significant limitation: the lack of a more detailed examination of the ascitic fluid. LP is a rare condition with no established diagnostic criteria, making its diagnosis quite challenging. In this context, some studies have emphasized the usefulness of ascitic fluid analysis in the diagnosis of LP. Some reports have suggested that LP exhibits an isolated increase in IL-6 levels (Watanabe et al. 2015), whereas PD-related peritonitis has been associated with elevations in IL-6 as well as other cytokines such as IL-8 (Ko et al. 1995). Monitoring various inflammatory cytokines in the ascitic fluid may be beneficial for the differential diagnosis. Regarding the SLE activity markers, some reports have indicated decreased complement levels and increased antinuclear antibodies in ascitic fluid from patients diagnosed with LP, which may be useful in the differentiation between LP and other forms of peritonitis (Schousboe et al. 1988; Watanabe et al. 2015). In this case, these parameters were not measured, and the inability to provide direct evidence of LP remains a significant limitation of this report. Future cases should involve proactive testing of these parameters to support the diagnosis.
In conclusion, unique autoimmune activation caused by COVID-19 can lead to unexpected pathological shifts and challenging diagnoses. Clinicians should consider various possible causes of immune flare-ups when managing autoimmune diseases in conjunction with COVID-19.
The authors declare no conflict of interest.
