Renal tubular acidosis complication of non-steroidal anti-inflammatory drugs induced interstitial nephritis and its complete resolution with steroids: A case report

Abstract

BACKGROUND

Non-steroidal anti-inflammatory drugs (NSAIDs) are extensively utilized for their analgesic and anti-inflammatory properties; however, their potential to induce renal adverse events remains a significant clinical concern. NSAID-induced interstitial nephritis (IN) and renal tubular acidosis (RTA) are rare but important complications of prolonged NSAID consumption. NSAID-induced RTA has been documented in limited case reports and usually resolves after stopping NSAIDs. However, our case shows a persistent, severe hypokalemia and hypophosphatemia linked to RTA from NSAID-induced IN, which resolved with steroid therapy.

CASE SUMMARY

A 70-year-old woman with a history of chronic NSAID use presented with weakness and palpitations. She had atrial fibrillation, severe hypokalemia (K 1.5 mmol/L), non-anion gap metabolic acidosis (HCO₃ 11 mmol/L, anion gap 11), hypophosphatemia (1.2 mg/dL), acute kidney injury (creatinine 2 mg/dL from 0.6), and rhabdomyolysis (creatine kinase 15620 U/L). Urinalysis showed pH 6.5, proteinuria, sterile pyuria, and a positive urine anion gap, indicating RTA. Electrolytes improved with supplementation and spironolactone, but didn’t normalize. At 4 weeks post-hospitalization, she still required supplementation; creatinine remained high. Renal biopsy revealed interstitial inflammation with active tubulitis, moderate interstitial fibrosis, and tubular atrophy. She was treated with 4 weeks of prednisone, leading to normalized electrolytes and improved creatinine (1 mg/dL), and the improvements were sustained 6 months post-treatment.

CONCLUSION

Our patient exhibited persistent metabolic disturbances even after the cessation of NSAID therapy. Despite moderate chronicity, active interstitial inflammation responded favorably to corticosteroid treatment, resulting in improved renal function and restoration of tubular function, with normalization of acid-base and electrolyte parameters.

Key Words: Interstitial nephritis; Non-steroidal anti-inflammatory drugs; Renal tubular acidosis; Hypokalemia; Steroids; Tubular dysfunction; Case report

Core Tip: This article underscores the importance of maintaining a high level of clinical suspicion for interstitial nephritis (IN) in case of non-steroidal anti-inflammatory drugs (NSAID)-induced persistent renal injury and dyselectrolytemia suggestive of renal tubular acidosis, and the critical role of biopsy in NSAID-induced IN. Role of steroids in drug-induced IN is debated; some study suggests their role only in early cases before fibrosis ensues. Our article emphasizes the potential role of steroids in persistent renal impairment and metabolic disturbances, even after discontinuation of NSAIDs, provided there is evidence of active interstitial inflammation and tubulitis on renal biopsy.

INTRODUCTION

Interstitial nephritis (IN) is an immune-related kidney injury often caused by drugs, infections, or autoimmune disorders[1]. Drug-induced interstitial nephritis (DI-IN) significantly contributes to acute and chronic kidney problems, marked by inflammation of the renal interstitium and tubular damage. Acute interstitial nephritis (AIN) occurs in 0.5%-3% of all kidney biopsies, and in 5%-27% of biopsies for acute kidney injury[1].

Drug-induced AIN (DI-AIN) typically develops within days to weeks after exposure to offending agents such as antibiotics (e.g., β-lactams, rifampin), proton-pump inhibitors, non-steroidal anti-inflammatory drugs (NSAIDs), and certain antivirals[2]. DI-AIN should be distinguished from drug-induced nephrotoxic acute tubular necrosis due to their different pathophysiology and treatment approaches. AIN, a key cause of acute renal failure, is a CD4⁺ T-cell-mediated reaction that damages tubules but spares glomeruli and blood vessels[3,4]. Antibiotics account for one-third of drug-induced cases. 1%-5% of NSAID users are at risk for DI-IN due to its widespread use[5].

NSAIDs can cause kidney injury by inhibiting prostaglandins that maintain renal blood flow, especially in a low effective circulatory volume state. The most common form of NSAIDs-induced renal injury is hemodynamically mediated acute tubular necrosis (ATN) resulting from afferent arteriolar vasoconstriction, reduced glomerular filtration, and ischemic tubular injury. They may also cause acute IN, often presenting with subnephrotic-range proteinuria and occasionally sterile pyuria[6]. Prolonged exposure to NSAIDs can trigger cytokine release (such as transforming growth factor-beta, interleukin-1, interleukin-4, insulin-like growth factor 1, endothelin-1), leading to progressive interstitial fibrosis and extracellular matrix buildup, resulting in tubular architectural destruction and irreversible loss of nephron mass, known as chronic IN[7].

As tubular cells are involved in solute and acid-base handling, AIN often causes tubular dysfunction, including impaired concentrating ability (polyuria), salt wasting, glycosuria with normal blood sugar, and aminoaciduria, collectively called Fanconi-like features when proximal tubules are affected. NSAIDs-induced interstitial inflammation may impair tubular function, especially hydrogen and bicarbonate handling, leading to renal tubular acidosis (RTA), most often distal (type 1) RTA due to impaired hydrogen secretion, though proximal (type 2) RTA may occur with significant proximal tubular injury[8] and often associated with severe electrolyte and acid-base disturbances, including hypokalemia which occasionally could be fatal[9,10]. There is limited published literature on NSAID-induced RTA, but most of these RTAs resolved with withdrawal of the offending agent[10]. We present a case involving NSAID-induced IN with resultant RTA and life-threatening hypokalemia, hypophosphatemia, and rhabdomyolysis, which entirely resolved with steroid treatment.

CASE PRESENTATION

Chief complaints

A 70-year-old Caucasian female presented with progressive worsening of generalized weakness and palpitations over several days.

History of present illness

She reported muscle soreness and chronic nausea due to gastroparesis but no other gastrointestinal symptoms such as vomiting or diarrhea. She denied chest pain, dyspnea, and urinary symptoms. She has no prior history of renal insufficiency. No recent hospitalization. Her medications include 600 mg ibuprofen and 440 mg naproxen 2-3 times a day for chronic headache, and occasional ondansetron; she does not use proton-pump inhibitors, herbal agents, or supplements.

History of past illness

Gastroparesis, tardive dyskinesia, and chronic headache.

Personal and family history

Noncontributory to current presentation.

Physical examination

On admission, the patient was afebrile, with a pulse of 92, blood pressure of 126/78 mmHg, respiratory rate of 18, oxygen saturation at 96% on room air, and a body weight of 55 kg. She was not in apparent distress. Examination of the head revealed dry mucous membranes; pupils were equal, round, and reactive to light and accommodation, and extraocular movements were intact. There was no jugular venous distention, thyromegaly, or lymphadenopathy noted in the neck. The cardiovascular exam revealed normal S1 and S2, but the rhythm was irregularly irregular without any murmurs, rubs, or gallops. On chest examination, bilateral air entry was equal, and no wheezes or rales were heard. The abdomen was soft, nontender, and nondistended, and bowel sounds were present. Extremities showed no cyanosis, clubbing, jaundice, or dependent edema. Musculoskeletal assessment revealed no restriction in the range of motion but generalized muscle tenderness over both extremities. The skin showed no pallor, cyanosis, or icterus, but slightly decreased turgor was noted. Neurologically, the patient was alert and oriented to person, place, and time, with no focal deficits; motor and sensory functions were intact, deep tendon reflexes were preserved, and cranial nerves II through XII were normal.

Laboratory examinations

Laboratory studies revealed a serum potassium level of 1.5 mmol/L, a bicarbonate level of 11 mmol/L, an anion gap of 11, a blood urea nitrogen of 23, a creatinine level of 2 mg/dL, a phosphorus level of 1.2 mg/dL, and a creatinine kinase (CK) level of 15620 U/L. Urinalysis showed a specific gravity of 1.015, 1+ protein, trace amounts of blood and leukocyte esterase, and negative nitrites. Urine microscopy identified 10-20 white blood cells per high-power field (hpf), 0-2 red blood cells/hpf, and no bacteria. The spot urine protein-creatinine ratio (UPCR) was 1.9 g/g (Table 1). As noted earlier, she has no history of chronic renal insufficiency, and her baseline creatinine was approximately 0.6 mg/dL four months prior to presentation.

Table 1 Laboratory work-up during admission and clinic follow-up.

		On admission	On discharge	4-week follow-up (pre-biopsy)	Post-steroid therapy	3-months f/up	6-months f/up	Reference range	Unit
Serum chemistry	Sodium	140	135	142	139	137	137	135-145	mmol/L
	Potassium	1.5	4.2	3	4.4	3.7	3.9	3.6-5.1	mmol/L
	Chloride	118	108	115	107	105	105	96-110	mmol/L
	Bicarbonate	11	22	20	27	24	25	22-28	mmol/L
	Phosphorus	1.2	2.8	1.8	3.7	3.1	3.4	2.5-4.8	mg/dL
	Anion gap	11	5	7	5	8	7	8-12	mmol/L
	Creatinine	2	1.5	1.2	1.0	1.0	1.0	0.5-1.1	mg/dL
	eGFR (MDRD)	32	37	45	55	55	55	> 90	mL/minute/1.73 m2
Urine chemistry	UPCR	1.9		0.6	0.2	0.1	0.1	< 0.03	g/g
	Sodium	58						20-110	mmol/L
	Potassium	36						12-62	mmol/L
	Chloride	86						55-125	mmol/L
Urine	UA	Specific gravity 1.015, pH 6.5, 1+ protein, trace blood, trace leukocyte esterase
	Microscopy	10-20 WBCs, 0-2 RBCs, no bacteria and no casts (per high power field)

eGFR: Estimated glomerular filtration rate; MDRD: Modification of diet in renal disease; UPCR: Urine protein creatinine ratio; WBC: White blood cells; UA: Urinalysis.

Imaging examinations

Renal ultrasound showed decent-sized kidneys bilaterally with slightly increased cortical echogenicity. Cortical thickness and cortico-medullary differentiation are relatively preserved. No nephrolithiasis or hydronephrosis observed.

MULTIDISCIPLINARY EXPERT CONSULTATION

Her acute kidney injury was presumed to be due to pre-renal etiology vs acute tubular necrosis due to NSAIDs exposure and rhabdomyolysis. NSAIDs were discontinued, and she was started on intravenous volume expansion with a balanced crystalloid solution along with aggressive intravenous and oral supplementation of potassium chloride, potassium phosphate, and oral sodium bicarbonate. Her renal function and CK started to improve. However, despite receiving round-the-clock electrolyte supplementation, she remained hypokalemic, hypophosphatemic, and acidotic even after 2 days. A nephrology consultation was called.

Given her normal anion-gap metabolic acidosis, urine electrolyte analysis was performed. The results showed a urine sodium of 58 mmol/L, urine potassium of 36 mmol/L, and urine creatinine of 86 mmol/L, with a urine anion gap of +8. Considering the presence of normal anion gap metabolic acidosis, positive urinary anion gap, severe hypokalemia, and hypophosphatemia, a diagnosis of RTA was suspected. Serologic testing, including anti-neutrophil cytoplasmic antibodies, anti-double-stranded DNA, anti-SS-A, and anti-SS-B, was conducted; all results were unrevealing. Serum protein electrophoresis did not reveal any monoclonal band, and the serum free light chain ratio (kappa/Lambda) was 1.38 (Table 2). She was started on spironolactone 25 mg twice daily, potassium chloride 40 mEq four times daily, Neutra-Phos two packets (8 mmol of elemental phosphorus in each packet) three times daily, and sodium bicarbonate 1300 mg three times daily. After six days of hospitalization, her electrolyte levels improved: Serum potassium reached 4.2 mmol/L, bicarbonate was 22 mmol/L, and phosphorus increased to 2.8 mg/dL. Spironolactone was discontinued, and she continued on potassium chloride 20 mEq twice daily, one packet of Neutra-Phos 1 packet twice daily, and sodium bicarbonate 650 mg twice daily. Her creatinine improved to 1.5 mg/dL. She was advised to strictly avoid NSAIDs.

Table 2 Serologic, infectious, and paraproteinemia work-up.

	Test	Result	Reference range	Unit
Serologic work-up	ANA	Negative	Negative
	Anti-dsDNA	4	< 10	IU/mL
	SS-A Ab	< 0.2	≤ 1.0	AI
	SS-B Ab	< 0.2	≤ 1.0	AI
	ANCA panel IFA	None detected
	Anti-GBM Ab, IgG	Negative
	RF	< 10	< 15	IU/mL
	C3	124	90-180	mg/dL
	C4	26	15-40	mg/dL
Monoclonal work-up	Serum protein electrophoresis	No monoclonal band
	Kappa/Lamda ratio	1.38	0.26-1.65
Infectious work-up	HIV ELISA	Negative
	Hepatitis panel	Negative hepatitis B surface antigen, & hepatitis C antibody

ANA: Anti-nuclear antibody; Anti-dsDNA: Anti-double-stranded DNA; SS-A: Sjögren’s syndrome antibody A/anti-Ro antibody; SS-B: Sjögren’s syndrome antibody B/anti-La antibody; ANCA panel: Anti-neutrophil cytoplasmic antibody; IFA: Immunofluorescent antibody; ant-GBM: Anti-glomerular basement membrane antibody; RF: Rheumatoid factor; C3: Complement factor 3; C4: Complement factor 4; HIV: Human immunodeficiency virus.

At the 4-week nephrology clinic follow-up, routine serum chemistry revealed worsening hypokalemia (potassium 3 mmol/L), acidosis (bicarbonate 20 mmol/L), and hypophosphatemia (phosphorus 1.8 mg/dL) (Table 1). She reported strict adherence to her electrolyte supplementation regimen and declined any further NSAID use since her hospitalization. Her creatinine level improved to 1.2, and the estimated glomerular filtration rate (eGFR) was 45 according to the MDRD equation. Repeat urine analysis showed persistent proteinuria and sterile pyuria, although the UPCR improved to 0.6 g/g. Due to concerns about IN, a computed tomography-guided renal biopsy was performed after shared decision-making. Repeat autoimmune serologies were ordered and showed no abnormalities.

Renal biopsy pathology revealed moderate interstitial inflammation with infiltration of lymphocytes, some plasma cells, rare neutrophils, and eosinophils, and mild-moderate tubulitis (Figure 1A and B). Moderate interstitial fibrosis and tubular atrophy (IFTA) were noted involving approximately 40% of the biopsied cortex (Figure 1C). Immunofluorescence and electron microscopy were unrevealing.

Figure 1 Light microscopic picture. A: Showing active interstitial inflammation with lymphocytes, plasma cells, rare neutrophils, and eosinophils (arrows); B: Showing mild to moderate tubulitis (blue arrows) and tubular atrophy with narrowed lumens in shrunken tubules and cuboidal epithelial cells (black arrows); C: Showing moderate interstitial fibrosis and tubular atrophy (pale blue arrows) involving approximately 40% of the biopsied cortex. Also has multifocal interstitial infiltrate (orange arrows).

FINAL DIAGNOSIS

NSAID-induced IN and distal RTA.

TREATMENT

She was subsequently started on prednisone 40 mg daily for 2 weeks with a quick taper over the next 2 weeks. After steroid therapy, renal function, electrolytes, and acidosis improved- creatinine was 1 mg/dL with eGFR of 55, potassium 4.4 mmol/L, phosphorus 3.7 mmol/L, and bicarbonate 27 mmol/L. All electrolyte supplementations were stopped.

OUTCOME AND FOLLOW-UP

During clinic follow-ups at three and six months after completing steroid therapy, creatinine and electrolytes remained stable (Table 1). Her residual renal insufficiency is presumably from moderate IFTA observed on renal biopsy, likely attributable to chronic heavy NSAID usage. Serum potassium and bicarbonate levels on admission and throughout the course, along with the supplementation dosage, are depicted in Figure 2.

Figure 2 Serum potassium and bicarbonate levels on admission and throughout the course, along with the supplementation dosage. A: Clinical course of serum potassium along with the dose of potassium chloride supplementation; B: Clinical course of serum bicarbonate along with the dose of sodium bicarbonate supplementation.

DISCUSSION

NSAIDs-induced RTA with different electrolyte and acid-base imbalances

In NSAID-induced distal RTA, urinary acidification is impaired by reduced H⁺ secretion by type A intercalated cells in the collecting duct, primarily through prostaglandin inhibition affecting the H⁺-ATPase pump[11]. The urinary anion gap [UAG = (Na⁺ + K⁺) - Cl^-] is consistently positive[12]. In distal RTA, impaired hydrogen ion excretion promotes potassium loss in exchange for sodium, and impaired H-K-ATPase pump activity reduces distal potassium reabsorption, leading to hypokalemia[9].

The mechanism of NSAID-induced proximal RTA is not well defined but appears to be due to carbonic anhydrase inhibition, an enzyme crucial for acid-base balance in renal tubules that facilitates the interconversion of carbonic acid to CO₂[13]. In proximal RTA, the UAG varies with disease phase: Early, when bicarbonate wasting is prominent, and serum bicarbonate is near-normal, the UAG is positive; later, as bicarbonate declines and serum acidifies, NH₄⁺ excretion increases, urinary chloride rises, and the UAG becomes negative, showing a biphasic pattern[8]. In contrast to distal RTA, in proximal RTA, impaired proximal sodium reabsorption leads to increased distal sodium delivery, secondary hyperaldosteronism, and increased kaliuresis.

In our patient, UAG was positive. She experienced severe metabolic acidosis, with a serum bicarbonate level of 11; however, the acidosis improved with a modest dose of alkali supplementation. In proximal RTA, although the acidosis is typically only mild-moderate, the required alkali load is significantly higher than that in distal RTA, often reaching up to 20 mEq/kg/day. Additionally, the initial urine pH was 6.5, which is inconsistent with a steady-state proximal RTA in the absence of alkali therapy, as urine pH in such cases is usually less than 5.5. She indeed experienced severe hypophosphatemia, potentially accompanied by phosphaturia, which is traditionally regarded as an indicator of proximal tubular dysfunction. However, persistent metabolic acidosis in distal RTA may cause proximal tubular phosphate reabsorption defects by suppressing NaPi-IIa and NaPi-IIc cotransporter activity, leading to decreased reabsorption and increased urinary phosphate loss. Chronic acidosis also promotes bone buffering, releasing calcium and phosphate into the bloodstream[14], which further increases urinary phosphate excretion. Together, reduced tubular reabsorption and increased filtered phosphate contribute to hypophosphatemia, poor bone mineralization, and osteomalacia or rickets in untreated distal RTA[8].

Management of drug-induced interstitial nephritis

Early withdrawal of the offending agent remains the cornerstone of treatment for suspected DI-AIN. However, when renal impairment doesn’t improve with medication withdrawal alone, steroids have been tried as a salvage option. The role of steroids in the treatment of DI-AIN is controversial. Although data appear ample, they have never definitively proven the benefits of steroids without reasonable doubt. There are very few high-quality randomized controlled trials (RCTs). Most evidence comes from retrospective cohorts or case series, which are prone to bias (selection bias, confounding) and protocol heterogeneity.

A 2018 systematic review of 8 retrospective studies with 430 patients (300 treated with steroids vs 130 without) reported limited evidence to support the use of corticosteroids in the treatment of DI-AIN; no meta-analysis was performed due to heterogeneity[15]. A subsequent retrospective cohort of 187 biopsy-proven AIN patients (158 treated with steroids, 29 untreated) found that steroid-treated patients had significantly better eGFR at follow-up (median 43 mL/minute vs 24 mL/minute) and lower dialysis-dependency at 6 months (3.2% vs 20.6%) and 24 months (5.1% vs 24.1%)[16]. A smaller randomized trial (n = 31) comparing oral prednisolone 1 mg/kg vs IV pulse methylprednisolone 30 mg/kg × 3 days, then oral, found no significant difference in the rate of complete response (approximately 56% vs approximately 60%)[17]. A 2025 systematic review of 23 studies (3 RCTs, 4 case series, 16 retrospective cohort studies); meta-analysis of 3 RCTs (using prednisolone at 1 mg/kg/day for ≥ 2 weeks) found a pooled effect size (Cohen’s d) of -2.24 (95%CI: -2.82 to -1.67), suggesting a significant improvement in serum creatinine with steroid use[18]. A ‘Columbia cohort’ of 139 patients with biopsy-proven AIN found that steroid therapy reduces serum creatinine (-2.3 mg/dL; 95%CI: -3.6 to -1.1) and dialysis-dependency at 6 months (11% vs 54%). In multivariate analysis, glucocorticoid therapy was an independent predictor of improved kidney function (adjusted difference in serum creatinine of -1.47 mg/dL; 95%CI: -2.68 to -0.27)[19].

Based on the above evidence, it appears that steroid therapy improves renal recovery in AIN. The magnitude of benefit is difficult to quantify reliably because of heterogeneity in the cause (drug vs other), timing of diagnosis, steroid dosing/regimen, follow-up duration, and lack of control groups. Most studies have used prednisone at a dose of 1 mg/kg/day for at least 2 weeks, with a gradual taper over another 2 weeks, and two small Indian RCTs found no benefit of pulse IV methylprednisolone over an oral steroid regimen[17,20].

The timing of steroid treatment is also of paramount importance. In 2008, a multicenter retrospective trial of 61 patients with DI-AIN, of whom 52 were treated with steroids, showed that steroid-treated patients had significantly lower final serum creatinine levels, and that earlier initiation of steroids correlated with better renal recovery[21]. Another retrospective Indian study of 83 patients with drug-induced AIN treated with steroids found that about 47% achieved “complete response” (serum creatinine < 1.5 mg/dL at 1 year)[22]. Worse outcomes are typically associated with interstitial fibrosis on biopsy[23]. This suggests that early initiation of steroid therapy, before interstitial fibrosis and tubular atrophy set in, may confer greater benefit. Some key recent data on steroids in IN are summarized below (Table 3).

Table 3 Key recent trials on steroids in interstitial nephritis.

Ref.	Design/N	Patient population	Steroid regimen	Key outcome(s)	Notes/limitations
Systematic review & meta-analysis
Quinto et al[15], 2019	Systematic review, 8 retrospective studies, 430 patients (approximately 300 steroid vs approximately 130 non-steroid)	Drug-induced AIN	Prednisone approximately 40-60 mg daily (5 studies); IV methylprednisolone approximately 1 mg/kg in two studies; duration 1.5-12 weeks	Mixed: Four studies showed benefit of steroids on SCrt, four did not. No meta-analysis due to heterogeneity	High risk of bias. Retrospective. Comparator arms poorly defined
Yu et al[18], 2025	Systematic review, 3 RCT, 4 case series, 16 retrospective case series (total 1205 patients; 952 received steroid treatment)	Biopsy-proven AIN (not necessarily exclusively drug-induced)	Oral prednisolone 1 mg/kg/day with slow taper over several weeks or IV methylprednisolone pulse therapy for 2 days to 3 days before oral prednisolone. One study used MMF as steroid sparing agent	Mixed: Eight studies showed improved renal function with steroids, whereas another eight found no significant difference. Meta-analysis of 3 RCT: Overall effect size is -2.24, with a confidence interval of -2.82 to -1.67	High risk of bias. Retrospective. Adverse effects of steroid is probably under-reported
RCTs
Ramachandran et al[20], 2015	Randomized (though small) trial. n = 29 (16 oral pred vs 13 pulse + oral)	Biopsy-proven. DI-AIN	Group 1: Oral prednisolone 1 mg/kg/day × 3 weeks then taper. Group 2: Methylprednisolone IV 30 mg/kg/day × 3 days then prednisolone 1 mg/kg/day orally × 2 weeks, then taper	At 3 months: CR (eGFR ≥ 60 mL/minute/1.73 m²) in 50% (Group 1) vs 61% (Group 2); no statistically significant difference	Small N. Short follow-up period of 3 months. No control group. Limited generalizability
Chowdry et al[17], 2018	Randomized (though small) trial. n = 31 (16 oral pred vs 15 pulse + oral)	Biopsy-proven AIN	Group A: Oral pred 1 mg.kg/day × 2 weeks. Group B: IV methylprednisolone 30 mg/kg/day × 3 followed by oral pred 1 mg/kg/day × 2 weeks with 2 weeks taper	No statistical difference in outcome: Complete remission (eGFR ≥ 60 mL/minute/1.73 m²) 56.2% (Group A) vs 60% (Group B). Additional 44% in Group A & 40% in Group B achieved partial remission (improvement but eGFR < 60 mL/minute/1.73 m2)	Small N. No control group. Limited generalizability
Retrospective
González et al[21], 2008	Multicenter retrospective. n = 61 (52 steroid-treated)	Biopsy-proven DI-AIN	Steroids started within approximately 2 weeks vs later (approximately 34 days) - dose not fully standardized	Treated patients had significantly lower final serum creatinine; earlier initiation correlated with better recovery	Retrospective. Modest size. Timing confounded by other factors
Fernandez-Juarez et al[23], 2018	Retrospective (n = 182 from 13 centers	Biopsy-proven DI-AIN- all treated with steroids	Compared treatment durations: High-dose for 3 weeks vs > 8 weeks	At 6 months: Mean recovered GFR approximately 34 mL/minute/1.73 m²; longer high-dose (> 3 weeks) or > 8 weeks duration not associated with better recovery. Delay in steroid initiation and interstitial fibrosis > 50% on biopsy strongly predicted worse outcomes (OR for fibrosis approximately 8.7)	No control group. Retrospective. Mixed durations & dosage
Prendecki et al[16], 2017	Retrospective: n = 187 (158 treated with steroids, 29 untreated)	Biopsy-proven AIN (not necessarily exclusively drug-induced)	Oral prednisolone (and a few with IV)	At 24 months: Median eGFR in steroid group 43 mL/minute vs 24 mL/minute in untreated (P = 0.01). Dialysis-dependence by 6 months: 3.2% steroid vs 20.6% untreated (P = 0.0022). By 24 months: 5.1% vs 24.1% (P = 0.0019)	Non-randomized. Potential for selection bias (sicker patients might not have been treated)
Rodelo-Ceballos et al[19], 2025	Retrospective: n = 139 (101 treated with GC)	Biopsy-proven AIN (various etiologies)	GC therapy; earlier initiation (≤ 7 days) vs later	GC-treated group had significantly greater delta SCrt reduction (-2.3 mg/dL; 95%CI: -3.6 to -1.1) & lower permanent dialysis-dependence at 6 months (11% vs 54%) in treated group. Multivariate: GC therapy independent predictor of improved kidney function (adj SCrt: -1.47 mg/dL; 95%CI: -2.68 to -0.27)	Retrospective. Single center cohort. Possible residual confounding

AIN: Acute interstitial nephritis; DI-AIN: Drug-Induced acute interstitial nephritis; GC: Glucocorticoid; GFR: Glomerular filtration rate; eGFR: Estimated glomerular filtration rate; SCrt: Serum creatinine; OR: Odds ratio; RCT: Randomized controlled trial.

In this patient, renal biopsy demonstrated moderate IFTA, along with active interstitial inflammation and cellular infiltration. DI-AIN was considered to persist, contributing to tubular dysfunction characterized by electrolyte and acid-base disturbances. After four weeks of NSAID discontinuation, neither her RTA nor renal functional impairment resolved; thus, steroid therapy was initiated, resulting in complete resolution of RTA. Although serum creatinine improved, it did not return entirely to baseline, likely attributable to pre-existing moderate IFTA associated with chronic NSAID exposure.

CONCLUSION

Although ATN is the most common NSAID-related renal injury, NSAIDs can also cause acute and chronic IN. Maintaining a high index of clinical suspicion and low thresholds for renal biopsy are crucial. RTA (predominantly distal variety) is a rare complication of NSAID-induced IN. Discontinuing NSAIDs is the cornerstone of treatment. If renal and tubular dysfunction persists after NSAID cessation, steroids may be beneficial if renal biopsy shows active inflammation, even in the presence of moderate chronicity.

ACKNOWLEDGEMENTS

We thank our patient for allowing us to submit the case and our institutions for supporting our work.