Diagnosis and Characteristics of presentation of Tubulointerstitial Nephritis and Uveitis Syndrome during the COVID-2019 pandemic

Abstract

Purpose:

To compare the diagnosis and clinical features of tubulointerstitial nephritis and uveitis syndrome (TINU) before and during the COVID-19 pandemic.

Methods:

Retrospective chart review.

Results:

Before the COVID-19 pandemic (March 2017 to March 2019), 1/561 (0.18%) new patient was diagnosed with TINU. During the pandemic (March 2020 to March 2022), 15/581 (2.58%) new patients were diagnosed with TINU. We found a significant increase in TINU case during the pandemic while the number of total new patients did not differ significantly (P=0.0005). Before the pandemic, 2/3 (66.7%) patients showed posterior segment findings: disc edema (33.3%), multifocal chorioretinopathy (33.3%), and peripheral vascular leakage on fluorescein angiography (33.3%). During the pandemic, 13/15 (86.7%) patients had posterior involvement: disc edema (73.3%), chorioretinal scars (13.3%), disc leakage (85.7%), peripheral vascular leakage (85.7%).

Conclusion:

This is the first study reporting an increased number of TINU during the COVID-19 pandemic, and most patients demonstrated posterior segment findings. With most of the American population now exposed to COVID-19, a large multi-center epidemiological study would be helpful to investigate any association of COVID-19 disease or vaccination with TINU in recent years. We hope to increase the awareness of TINU among clinicians and improve the understanding of clinical features of TINU which are no longer limited to bilateral non-granulomatous anterior uveitis. With more study and data in the future, the uveitis characteristics of TINU may need to be modified.

Introduction

Tubulointerstitial nephritis and uveitis syndrome (TINU) is an uncommon entity characterized by interstitial nephritis and uveitis, which comprises approximately 0.2 to 2% of uveitis patients¹. Interstitial nephritis commonly manifests as renal dysfunction (elevated creatinine), abnormal urinalysis (elevated urine β2-microglobulin, low-grade proteinuria, eosinophiluria, white cell casts, pyuria), together with systemic illness (fever, malaise, anemia, elevated erythrocyte sedimentation rate). In the TINU diagnostic criteria proposed by Mandeville et al² in 2001, the “typical” uveitis was defined as bilateral anterior uveitis with or without intermediate or posterior uveitis, presenting within 2 months prior to or 12 months after an acute interstitial nephritis episode. Mandeville et al² described 80% of their TINU patients with anterior uveitis while only 20% with posterior or panuveitis². Reported posterior segment findings with TINU include macular edema, optic disc edema, chorioretinal lesions, multifocal choroiditis, choroidal neovascular membranes, neuroretinitis, acute posterior multifocal placoid pigment epitheliopathy and retinal pigment epithelial detachments^3–12. In a case series of 17 TINU patients published by Koreishi et. al¹³, optic disc edema was reported as the most common posterior finding, present in 41.2% of their TINU patients. Retinal vasculitis evident by leakage on fluorescein angiography has also been reported in small case series of TINU^14,15. It remains unclear whether the increased proportion of intermediate/posterior findings is secondary to a change in disease phenotype or increased detection with advanced multimodal imaging.

The pathogenesis of TINU is likely multifactorial, and a variety of triggers have been investigated. Genetic susceptibility associated with human leukocyte antigen, autoimmune predisposition, and drug- or infection- related etiologies have been postulated^2,16–18. Recently, a small case series reported sarcoid-like uveitis in 7 patients following COVID-19 infection, 4 of whom had findings of TIN post COVID-19 disease¹⁹. While no definitive association of COVID-19 infection or vaccination with TINU can be drawn from this small study, several uveitis specialists on the American Uveitis Society listserv reported possibly seeing more cases of TINU during the COVID-19 pandemic. Our study investigates the demographics, ocular findings, and treatment of TINU at one tertiary referral center with a detailed comparison of these factors before and during the COVID-19 pandemic.

Methods

A retrospective chart review was performed at a single tertiary referral center after obtaining Institutional Review Board approval. We reviewed baseline demographic data including age, sex, ethnicity, potential triggers related to COVID-19 vaccination or disease, as well as laboratory results and treatment modalities. Clinical exam and imaging findings including optical coherence tomography (OCT) and fluorescein angiography (FA) were analyzed. Diagnostic criteria of TINU included bilateral anterior uveitis and evidence of tubulointerstitial nephritis (TIN) with either (1) a positive renal biopsy, (2) an elevated urine β−2 microglobulin (ub2m) (normal 0–300 ug/L) or an elevated urine β−2 microglobulin to urine creatine ratio (normal 0–300 ug/g CRT), or (3) evidence of nephritis (elevated serum creatinine and/or abnormal urine analysis). Specific diagnostic evaluation was also performed to rule out other etiologies, including sarcoidosis, syphilis, tuberculosis and HLA-B27-associated uveitis. The association between categorical variables was assessed with Fisher’s exact tests. Mixed-effects models were not used due to small sample sizes. Five percent significance (P-value <0.05) was used for all tests.

Results

The rate of TINU diagnosis before and during COVID-19 Pandemic (Table 1)

Table 1.

Incidence of TINU before and during COVID-19 Pandemic

	March 2017 to March 2019	March 2020 to March 2022	P Value
Number of patients with diagnosis of TINU	3	15
Number of new patients diagnosed with TINU (NTINU)	1	15
New Patients (NP)	561	581
NTINU/NP (%)	0.18	2.58	0.0005

There was no overlap of patients with TINU identified in the two different time frames. The association between categorical variables was assessed with Fisher’s exact tests. Five percent significance (P-value <0.05) was used.

In our tertiary university hospital, urinalysis, comprehensive metabolic panel, and ub2m were ordered by either the nephrology or ophthalmology department as initial screening tests for TINU. If ub2m was normal, the nephrology department determined whether a kidney biopsy was needed based on other test results including renal ultrasound, urinalysis, and comprehensive metabolic panel. We identified all the patients with the ICD-10 diagnosis code of TINU in the electronic medical record (EMR) between March 2017 to March 2019 (pre-pandemic) and between March 2020 to March 2022 (pandemic period). Some patients may not carry the ICD-10 diagnosis code of TINU in the EMR system. To avoid missing potential TINU patients, we also screened all patients with ub2m testing in the selected timeframes and examined their final diagnosis via a comprehensive chart review.

Between March 2017 to March 2019, we treated 3 patients with TINU. Among these 3 patients, only 1 patient was newly diagnosed with TINU. This represented 0.18% of new patients among the 561 new patients during this timeframe. In contrast, we diagnosed 15 new patients with TINU among the 581 total new patients between March 2020 to March 2022, which represented 2.58% of new patients. All three TINU patients treated in the pre-pandemic period moved out of the state and no longer followed at our institution. Therefore, there was no overlap of patients with TINU identified in the two different timeframes. We observed a significant increase in new TINU cases while the number of new patients did not change significantly during the COVID-19 pandemic compared to the pre-pandemic period (Table 1. P=0.0005).

Demographics, laboratory evaluation and treatments (Table 2)

Table 2.

Demographic, laboratory evaluations and treatment.

Patient	Sex	Race	Age	Urine β−2 microglobulin ug/L (nl 0–300 ug/L)	Renal biopsy	COVID vaccine or disease	Oral Steroid	IMT
March 2017 to March 2019
1	F	C	9	36492	+ TIN	NA	+	MTX
2	F	C	15	1286	-	NA	+	MTX
3	M	C	6	317	-	NA	+	MTX and then MMF
March 2020 to March 2022
4	F	C	13	379	-	V (AODO)	-	MMF and then MTX
5	M	H	16	1298	-	-	+	MMF+ADA
6	M	H	7	783	-	-	+	MTX+INF
7	F	C	12	8822	+ TIN	V, D (AODO)	+	MMF+ADA
8	M	C	15	3733	+ Non-specific fibrosis and scarring	V (AODO)	+	MMF+INF
9	F	C	10	3680	-	-	+	MMF+INF
10	M	C	12	658	-	V (AODO)	+	MMF+ADA
11	F	C	14	682	-	-	+	MMF (INF requested but LTFU)
12	M	C	9	259 (55 at 1 month) * B2M/Cr 301 (nl 0–300 ug/g)	-	-	+	MTX+ADA
13	M	H	13	331	-	-	+	Started MMF but LTFU
14	M	C	13	419	-	V (1 Mo BODO)	+	MMF
15	F	H	10	47 * B2M/Cr 362 (nl 0–300 ug/g)	-	D (AODO)	+	MMF and then MTX
16	M	C	11	816	-	-	+	MMF
17	M	C	19	3775	-	V (6 Mo BODO)	+	MMF
18	F	C	34	224 (on oral prednisone) * WBC cast in urine	-	V (ADOD)	+	MMF

F, female; M, male; C. Caucasian; H, Hispanic.

nl: normal limit. WBC, white blood cell. B2M/Cr, urine β−2 microglobulin to urine creatinine ratio.

NA, not applicable. V, vaccine. D, disease. AODO, after ocular disease onset. BODO, before ocular disease onset. Mo, month.

TIN, tubulointerstitial nephritis.

IMT, immunomodulatory therapy. MTX, methotrexate; MMF, mycophenolate; ADA, adalimumab; INF, infliximab. LTFU, lost to follow up.

Before the COVID-19 pandemic between March 2017 to March 2019, all 3 patients with diagnosis of TINU were Caucasian. The mean age of presentation was 10 (range 6–15) and 2 (66.7%) patients were female. Ub2m was elevated in all patients (mean value 12698 ug/L. Normal 0–300 ug/L). Renal biopsy was performed in 1 patient (Patient 1) with sterile pyuria and proteinuria which confirmed inactive interstitial nephritis and extensive chronic inflammation. All 3 patients required oral corticosteroids in addition to topical corticosteroid eyedrops. Later, each patient was treated with steroid-sparing immunomodulatory therapy (IMT), only requiring antimetabolites. These patients did not require tumor necrosis factor inhibitors for disease control.

During the COVID-19 pandemic between March 2020 to March 2022, 11 (73.3%) patients were Caucasian, and 4 (26.7%) patients were Hispanic. The mean age of presentation was 13.9 (range 7–34) and 6 (40%) patients were female. Patients 14 and 17 had onset of ocular symptoms within 1 month and 6 months of COVID-19 vaccination, respectively. All other patients either received COVID-19 vaccines or contracted the disease after the onset of uveitis. Ub2m was tested in all patients (mean value 1727 ug/L) and was elevated in 12 (80%) patients (normal 0–300 ug/L). Patient 12 initially had high normal ub2m 259 ug/L (normal 0–300 ug/L) and borderline elevated ub2m/urine creatinine ratio 301 ug/g CRT (normal 0–300 ug/g CRT). We suspected that Patient 12 might be in the phase in which the ub2m started to normalize. In 4 weeks, ub2m dropped to 55 ug/L (dropped by 4.7-fold) and the ub2m/urine creatinine ratio normalized to 196 ug/g CRT. This patient was evaluated by a nephologist who agreed this was consistent with a mild case of TIN and no renal biopsy was needed. Patient 15 had normal ub2m 47 ug/L (normal 0–300 ug/L) but an elevated urine β−2 microglobulin/urine creatinine ratio 362 ug/g CRT (normal 0–300 ug/g CRT) after corrected for the urine creatinine. Patient 18 initially had abnormal urinalysis with white blood cell casts. The u2bm was ordered but not done until the patient had been on 20 mg daily oral prednisone for 2 weeks. By that time, the u2bm was normal 224 ug/g, but likely due to the treatment. Based on the abnormal urinalysis, we think she met the criteria for TIN. Renal biopsy was performed in 2 patients. Patient 7 had chronic active tubulointerstitial nephritis and Patient 8 had non-specific post-inflammatory fibrosis and scarring. Fourteen (93.3%) patients received oral corticosteroids in addition to topical corticosteroid eyedrops and 15 (100%) patients were started on IMT. Seven (46.7%) patients were treated with antimetabolites alone while 8 (53.3%) patients required tumor necrosis factor inhibitors (adalimumab or infliximab) in addition to antimetabolites. No significant difference in the number of patients requiring tumour necrosis factor inhibitors (P=0.2157) was found before and during the COVID-19 pandemic.

Clinical findings (Table 3)

Table 3.

Clinical findings.

Patient	AC	KP	PS	VC	Snowballs	Disc Edema	Chorioretinal Scars	Disc Leakage on FA	Peripheral Retinal Vascular Leakage	CME on OCT
Between March 2017 to March 2019
1	+	-	+	+	-	+	+	ND	ND	ND
2	+	-	-	+	-	-	-	ND	ND	ND
3	+	-	-	+	+	-	-	-	+	-
Total	3/3	0/3	1/3	3/3	1/3	1/3	1/3	0/1	1/1	0/1
Between March 2020 to March 2022
4	+	-	-	+	+	-	-	-	+	-
5	+	-	-	+	-	-	-	+	+	-
6	+	-	-	+	+	-	-	ND	ND	-
7	+	+	-	+	-	+	-	+	+	-
8	+	+	+	-	-	+	-	ND	ND	-
9	+	+	+	+	-	+	-	ND	ND	-
10	+	-	-	+	-	+	-	ND	ND	ND
11	+	-	-	+	-	+	-	ND	ND	-
12	+	+	-	+	+	+	-	+	+	+
13	+	+	+	+	-	+	-	ND	ND	-
14	+	+	-	+	-	+	+	+	-	-
15	+	+	-	+	-	+	+	ND	ND	-
16	+	-	-	-	-	-	-	ND	ND	-
17	+	+	+	+	-	+	-	+	+	-
18	+	+	+	+	+	+	-	+	+	-
Total	15/15	9/15	5/15	13/15	4/15	11/15	2/15	6/7	6/7	1/14

AC, anterior cell. KP, keratic precipitates. PS, posterior synechiae.

VC, vitreous cell. FA, fluorescein angiography. CME, cystoid macular edema. OCT, optical coherence tomography. ND, not done.

Between March 2017 and March 2019, all 3 patients had non-granulomatous anterior segment inflammation, and posterior synechiae was seen in 1 patient (33.3%). Vitreous cell was seen in all 3 patients but only 1 (33.3%) patient had snowballs. Two (66.7%) patients had posterior segment findings on clinical exam or multimodal imaging, including disc edema (33.3%), multifocal chorioretinopathy (33.3%), or peripheral vascular leakage on FA (33.3%). Patient 1 had multifocal chorioretinopathy, which appeared to involve two types of choroidal lesions, the first characterized by large lesions scattered across the fundus beyond the arcade, and the second by small, punched-out chorioretinal scars inferiorly.

Between March 2020 and March 2022, all 15 patients (100%) presented with anterior chamber cells, 9 (60%) with non-granulomatous keratic precipitates and 5 (33.3%) with posterior synechiae. Thirteen (86.7%) patients were found to have vitreous cell while 4 (26.7%) had snowballs. Thirteen (86.7%) patients had various posterior segment findings identified by clinical exam or multimodal imaging. Eleven (73.3%) patients showed optic disc edema (Figure 1) while 2 (13.3%) had chorioretinal scars. FA was performed on 7 patients: six (85.7%) patients had disc leakage, and six (85.7%) demonstrated peripheral retinal vascular leakage without clinically visible vasculitis (Figure 2A-D). Fourteen patients had macular OCT which revealed cystoid macular edema in 1 (7.1%) patient. No significant difference in the number of patients presenting with optic disc edema (P=0.2451) or peripheral retinal leakage on FA (P>0.99) was found before and during the COVID-19 pandemic.

Figure 1.

Optic nerve OCT of Patient 7 showing bilateral thickened retinal nerve fiber layer due to optic disc edema.

Figure 2:

Ultra-wide field fundus photos and late-phase images of fluorescein angiography from Patient 7. A-B. Ultra-wide field fundus photos revealed bilateral optic disc edema without any chorioretinal lesions or scars in right eye (A) or left eye (B). No clinically visible vasculitis was seen in either eye. C-D. Images of the right eye at 10 minute 14 second (C) and the left eye at 10 minute 33 second (D) on fluorescein angiography demonstrated diffuse peripheral retinal vascular leakage without staining of the major vessels, and bilateral late disc leakage.

Discussion

TINU is a clinical diagnosis typically characterized by sudden onset bilateral non-granulomatous anterior uveitis, classically affecting patients less than 20 years of age, with a reported female predominance. The first diagnostic criteria for TINU were developed by Mandeville et al² in 2001, which described TINU as definite, probable, or possible. The clinical findings included renal dysfunction (elevated creatinine), abnormal urinalysis (urine β2-microglobulin, low-grade proteinuria, eosinophiluria, white cell casts, pyuria), and a variety of non-specific systemic illness (weight loss, anorexia, malaise, fatigue, rash, abdominal or flank pain, arthralgias, or myalgias). “Typical” uveitis for TINU was defined as bilateral anterior uveitis with or without intermediate or posterior uveitis, presenting within 2 months prior to or 12 months after acute interstitial nephritis (AIN). Definite TINU requires typical uveitis and AIN diagnosed with renal biopsy or clinical findings from all three diagnostic criteria: abnormal renal function, abnormal urinalysis and a systemic illness lasting ≥ 2 weeks. Probable TINU may have atypical uveitis with AIN diagnosed histopathologically, or typical uveitis with AIN diagnosed with incomplete clinical criteria. Based on these diagnostic criteria, 2 of our patients had definite TINU and 16 had probable TINU. Our patients were generally young with no clear female predilection, and one adult patient presented at age of 34.

In 2021, The Standardization of Uveitis Nomenclature (SUN) Working Group proposed updated diagnostic criterion of TINU²⁰, which were similar to those proposed by Mandeville et al, but simpler, eliminating the concepts of probable and possible TINU, and not including the non-specific characteristics of systemic illness. Key criteria included anterior chamber inflammation and evidence of tubulointerstitial nephritis with either 1) a positive renal biopsy or 2) evidence of nephritis (elevated serum creatinine and/or abnormal urine analysis) and an elevated urine β−2 microglobulin. As a tertiary referral center, many patients were referred to our institution after prolonged uveitis activity, resulting in delays from the acute presentation. Since the AIN component can be self-limited, laboratory markers may be normalized upon initial presentation. In a study by Goda et al¹⁶, only 25% of TINU patients had elevated creatinine at the time of evaluation, however, other studies showed that ub2m can remain elevated for months, occurring in 90–100% of TINU cases^21–23. Ub2m has been shown to be a reliable marker for TINU when other etiologies with similar findings are excluded (sarcoidosis, syphilis, tuberculosis and HLA-B27-associated uveitis )^21–23. Urine β−2 microglobulin to urine creatine ratio was not included in the initial diagnostic criteria made by Mandeville et al. However, this ratio is important for correcting ub2m level based on urine creatinine, which has been recommended by nephrology and also demonstrated to be a reliable indicator of renal dysfunction^24,25. Therefore, we use both ub2m level and urine β−2 microglobulin to urine creatine ratio as markers for renal dysfunction and abnormal urinalysis.

Several uveitis specialists on the American Uveitis Society listserv individually reported that they may be seeing more TINU patients during the COVID-19 pandemic, however, no formal study has been done to confirm this observation. We have now confirmed a significant increase of TINU cases during the COVID-19 pandemic while the number of new patients did not differ significantly. The cause of the increase remains unclear. In our series, we were unable to prove or disprove the association between TINU or COVID-19 infection/vaccination due to limited data. Only 2 of our patients had COVID-19 vaccines prior to the onset of ocular symptoms. All other patients during the pandemic either received COVID-19 vaccines or contracted COVID-19 disease after the onset of uveitis. In fact, many of our patients were not eligible for COVID-19 vaccines during the early pandemic due to young age given delayed authorizations of vaccines in pediatric population in the United States. In the United States, Pfizer-BioNTech and Moderna COVID-19 vaccines were first granted emergency use authorization for adults (age of 16 and above) on December 10, 2020 and December 17, 2020, respectively. The Pfizer-BioNTech vaccine was subsequently authorized for adolescents (aged 12–15) on May 10, 2021, and children (aged 5–11) on October 29, 2021. COVID-19 vaccines were approved for children (aged 6 month- to 5-year-old) on June 18, 2022. By now, most of the American population has either been vaccinated against COVID-19 or have contracted COVID-19 disease. With the majority of the population now exposed, a large multi-center epidemiological study would be helpful to investigate any association of COVID-19 disease or vaccination with TINU in recent years.

In contrast to the common thought that TINU is an isolated bilateral anterior uveitis, we identified a variety of intermediate and posterior manifestations in the majority of our patients. In current study, vitreous cell was seen in 88.9% of the patients and snowballs in 27.8%. With regard to posterior segment findings, optic disc edema was most common, occurring in 66.7% of our patients. This is consistent with the study by Koreishi et. al¹³ in which they reported disc edema as the most common posterior findings, present in 41.2% of their TINU patients. There have been reports of retinal vasculitis with leakage on FA in TINU patients, many of whom did not have clinically visible vasculitis. In the study of Yang, et al., FA was performed in 13 of 32 Chinese TINU patients, and peripheral vascular leakage was found in 22/26 eyes (84.62%)¹⁴. Cao et al. performed FA in 18 eyes of 9 TINU patients and found 13 eyes (72.2%) with peripheral vascular leakage, 5 eyes (27.7%) with optic disc leakage and 6 eyes (33.3%) with macular leakage¹⁵. Not every patient of ours had FA and OCT, but increased utilization of multimodal imaging has augmented our ability to detect posterior pathologies that are otherwise not easily visible on fundoscopy in TINU patients. A total of 8 patients in our series had FA (1 pre-pandemic and 7 during pandemic), 87.5% of whom had peripheral retinal vascular leakage and 87.5% had disc leakage. Our findings are consistent with recent studies, which demonstrate that posterior segment involvement is not uncommon in TINU and emphasize the importance of comprehensive examination and multimodal imaging. We did not observe any significant difference in the number of patients presenting with optic disc edema or peripheral retinal leakage on FA before and during the COVID-19 pandemic, although a small pre-pandemic sample size limits detailed analysis. We hope to increase the awareness of TINU among clinicians and improve the understanding of clinical features of TINU which are no longer limited to bilateral non-granulomatous anterior uveitis. With more study and data in the future, the uveitis characteristics of TINU may need to be modified.

In the study of Mandeville et al., topical corticosteroids alone were used in 21/123 (17%) of TINU patients, while systemic corticosteroids were used in 99/123 (80%) of the TINU patients, 13 (11%) of whom received it exclusively for ocular disease. In 11 of their patients, steroid-sparing IMTs were used to treat uveitis². More recent studies report a higher rate of IMT use, with 5/9 (56%) patients and 7/17 (41%) requiring IMT in studies by Sobolewski³ and Koreishi et al¹³, respectively. In our cohort, oral prednisone was used in 17/18 (94.4%) patients and IMT was recommended to all 18 patients. Ten (55.6%) of patients were treated with antimetabolites alone while 8 (44.4%) patients required addition of tumour necrosis factor inhibitors. Though biologics use was only seen in TINU patients during the COVID-19 pandemic, we did not identify a significant difference in the number of patients requiring biologics in the two different timeframes. However, our study is likely underpowered to detect a statistically significant difference, and a larger, long-term evaluation of TINU cases is necessary to reveal the impacts of different therapies on outcomes.

Conclusions

To our knowledge, this is the first study to formally evaluate TINU cases before and during the COVID-19 pandemic. We report a significant increase in TINU cases during the pandemic, though we cannot prove or disprove an association with COVID-19 infection or vaccination in our cohort due to limited data. Most patients demonstrated intermediate and posterior segment findings, especially with increased use of multimodal imaging. All patients in this case series were treated with steroid-sparing IMT which may reflect institutional practice pattern. Many of the patients required tumour necrosis factor inhibitors in addition to antimetabolites for uncontrolled ocular inflammation. We recommend a thorough fundus examination and use of multimodal imaging in patients with suspected TINU. Posterior findings may warrant more aggressive therapy, which may reduce long-term complications and improve visual outcomes.