In this study, we reviewed all patients with biopsy-proven AIN followed at a single Tunisian adult Nephrology center to identify the clinical, laboratory, etiological, and evolutionary characteristics.
In studies, AIN accounts for 0.5–6% of all renal biopsies. This prevalence ranges from 5–27% in biopsies conducted for investigating AKI. Our research found an overall prevalence of 2.84%, suggesting that AIN is less frequent in our center.
The incidence of AIN varies from year to year, experiencing a resurgence over the last decade [14, 18–20]. Some authors attribute the increased incidence of AIN to the rising number of biopsies performed in recent years and the increasing prescription of drugs that can cause immune-allergic interstitial damage [14, 21]. In our study, the peak incidence of AIN was observed between 2010 and 2013, subsequently decreasing over the past decade.
The mean age of our patients was 50.58 years. Our results are consistent with those documented in various studies, which show an average age range of 40 to 65 years [11, 13, 22, 23]. When examining the age groups, it was observed that the adult population, especially those aged 61 to 75 years, were most affected. This is consistent with similar observations in other studies [24, 25]. This pattern may be related to the higher medication use of these patients and their increased life expectancy.
In our study, 22.2% of patients experienced fever, with a notable link to drug-induced causes. Existing literature reports fever occurrence in 20–42% of cases and commonly link it to drug-induced AIN. The exact cause remains unidentified, emphasizing its nature as an epidemiological observation. A skin rash was observed in 19.4% of our patients, in 85.71% of cases, it was associated with immune-allergic AIN. This finding is consistent with other studies where the rate varies between 8% and 23% of cases [6, 25–30]. Only 13.88% of patients had oligoanuria at the time of nephropathy diagnosis. In the literature, the reported proportions of oligoanuria varied from 12–30%. In the series by Clarkson et al., oliguria was more common with a prevalence of 51%. Polyuria is not common in AIN; it is rather associated with chronic tubulointerstitial nephropathies. Its prevalence in our series was 5.54%, possibly explained by a urine concentration defect due to an associated tubular impairment.
Proteinuria during AIN is usually low, ranging from 0.4 g/d to 1.4 g/d [31–35], because interstitial nephropathy spares the glomeruli. Our results confirm this. Leukocyturia was present in 58.33% of our patients, it was aseptic in 80.95% of cases and indicative of infectious origin in 10% of patients. Our results align closely with the literature, where this rate ranges from 50% [36] to 85% [13, 32]. Among our patients, 25% exhibited microscopic hematuria during urine analysis. The prevalence of this condition varies, ranging from 30–70% across different studies [27, 33, 37]. Formerly, the presence of red blood cell casts was considered a specific indicator of glomerular disease [38]. Nevertheless, up to one-third of patients with AIN may display red blood cell casts, potentially due to interstitial blood vessel damage leading to erythrocyte leakage into the tubular lumen [38].
The gold standard for establishing a definitive diagnosis of AIN is renal biopsy [39–40]. However, in certain cases of immune-allergic AIN, this invasive procedure can be avoided and only recommended in the presence of an atypical presentation or lack of improvement after removal of the causal agent [21]. Several studies in the literature have demonstrated the diagnostic value of renal biopsy in cases of unknown origin of acute kidney injury [11, 24] Recently, the urinary level of CXCL9 has been validated as a biomarker for the diagnosis of AIN [1] and it has also shown that some AIN etiologies can be detected by Positron-Emission Tomography-Computed Tomography Imaging [41–43].
An inflammatory infiltrate in the renal interstitium is a key indicator of this kidney disease, predominantly located in the renal cortex [9, 44]. It consists mainly of T lymphocytes, monocytes, eosinophils, and other plasma cells [18, 45]. The distribution of the infiltrate and the dominant cell type varied in our study. Eosinophils (EOS) were found in only 5.55% of all biopsies in our study, they were associated with an immune-allergic etiology in 50% of cases. In the literature, the presence of EOS in histology ranges from 18% in Muriithi's series [13] to 94% in Clarkson's series [36], where the drug etiology was dominant. The presence of EOS in the inflammatory infiltrate in immune-allergic AIN suggests the involvement of allergic hypersensitivity mechanisms [46].
Histologic examination revealed interstitial fibrosis in 27.77% of patients. In fact, kidney damage during AIN progresses to fibrosis and this progression typically occurs within 7 to 10 days following the onset of the acute inflammatory process [18]. The development towards interstitial fibrosis and tubular atrophy can be prevented by avoiding the offending agent or promptly initiating steroid treatment [47]. Review of the histologic features of AIN in our series, led us to believe that renal biopsy remains an important tool in cases of AIN, because even if the diagnosis can be made by other means [1, 41–43], evaluation of the renal parenchyma remains important for prognostic purposes, given the high number of biopsies containing interstitial fibrosis and the high number of associated tubular lesions found in our patients.
Over the years, there has been a shift in etiology from the original description of AIN when infections were predominantly the main cause [48, 49]. At present, drugs are the main culprits in the majority of cases [4, 21, 50, 51]. Between 70 and 90% of all diagnosed cases of AIN are of immune-allergic origin [2, 4, 13, 21, 36, 52].
In our series, 41.66% of all patients had immune-allergic AIN. This scenario is further complicated by the increasing use of over-the-counter drugs such as NSAID or PPI [53]. Patients often perceive these drugs to be safe with minimal side effects and may use them for chronic conditions for extended periods, thereby increasing the risk of adverse effects [5, 36, 54–56]. Several studies suggest that the use of antibiotics is the main cause of immune-allergic AIN [14, 23, 33, 57, 58]. In our series, antibiotics were the second most common cause, accounting for 28.56% of cases, and NSAID were the most common cause of immune-allergic AIN. However, other authors reported that NSAID were the second most common cause after antibiotics in the etiology of immune-allergic AIN [13, 23, 25, 47, 59].
Apart from the immune-allergic mechanism of drug-associated AIN, our series included one patient who presented with an unusual drug-associated AIN. This was a renal sarcoid-like reaction associated with rituximab after r-CHOP treatment for mantle cell lymphoma [60].
In different studies, the frequency of infectious interstitial nephritis (IIN) ranged from 3–15% [24, 52, 61, 62]. In our study, IIN accounted for 13.88% of cases. It can occur as a result of direct renal invasion by microorganisms or the release of pro-inflammatory cytokines from a primary site of infection [14]. Bacterial pyelonephritis is the most common form and involves organisms such as Streptococcus, Staphylococcus, Salmonella and Brucella [3, 63]. Parasitic IIN associated with giardiasis has been reported [64] and we observed a severe case associated with dermatitis and hemolytic anemia in a patient with a triple parasitic infection, a first reported case [46].
In our series, 13.88% of patients had AIN secondary to systemic disease. The frequency of this etiology varies between studies from 1–24% [13, 26, 33, 52].
TINU, with a prevalence of 3.5 cases/million population [65], was diagnosed in only one patient in our series, whereas AIN was classified as idiopathic in 25% of cases. In the literature, idiopathic AIN represents 4% (6,21) to 20% [32]of AIN cases. Idiopathic cases in our series had no relevant known drug exposures and were classified as idiopathic after an investigative work-up. However, we believe that some of the idiopathic cases of AIN in our series are actually cases of TINU, since ocular abnormalities may appear later than renal damage [66]. Hence the importance of ophthalmological follow-up of patients with idiopathic AIN.
Concerning therapeutic management, there is a general agreement that discontinuation of the suspected causative medication is the primary approach in treating drug induced AIN [9, 19].
If kidney function does not show signs of recovery within 5–7 days after discontinuing the suspected medication, studies recommend initiating treatment with steroids [9]. Prednisolone is consistently described as the reference molecule. The initial dose varies from 0.5 mg/kg/day to 1 mg/kg/day depending on the series [23, 25, 33, 67–69]. However, intravenous methylprednisolone therapy at a dose of 500 mg/day for 3 days followed by oral corticosteroid therapy may be proposed [18, 47]. From another perspective, some studies suggest the use of other immunosuppressive agents such as mycophenolate mofetil, especially in the setting of corticosteroid failure [70]. In view of the rapid onset of interstitial fibrosis in AIN, as demonstrated in this series, we recommend early initiation of corticosteroid therapy in this etiology of AIN within no more than 7 days of diagnosis, since discontinuation of the drug alone would not be sufficient to repair the histological damage.
Our management aligns with the literature data where corticosteroid therapy represents the primary treatment for idiopathic interstitial nephritis [6, 33, 71]. Few studies suggest immunosuppressants as the first alternative [70].
The standard protocol for TINU involves systemic corticosteroid therapy, typically using prednisolone at a dosage of 1 to 1.5 mg/kg/day for a 4-week period, with possible extensions based on the patient's initial response. Specialized management of ophthalmological complications is also essential for cases that are recurrent or refractory [34, 72, 73]. Corticosteroid therapy is the mainstay treatment for interstitial nephritis associated with systemic diseases, supported by evidence of its effectiveness in managing the condition [13, 34, 71]. In cases of resistance to corticosteroids, alternative immunosuppressive treatments like rituximab or mycophenolate mofetil may be considered.
In our series, the median follow-up was 2.2 years, in line with the median follow-up found in the literature, which varies between two and three years in the majority of studies [13, 23, 26, 32, 33, 47, 71].
The course of renal function in AIN is often favorable. The time to renal recovery in AIN is often described in the literature as highly variable. Indeed, recovery can take from two weeks to three months. In our series, the evolution of renal function was generally favorable. Twenty patients, or 71.42% of cases, achieved renal recovery. Our results are similar to those reported in the literature (Table 5).
Table 5
Renal recovery in the literature and in our series
| Series | Total recovery of renal function | Partial recovery of renal function | No recovery of renal function |
| Fernandez et al. (28) | 41,2% | 45,6% | 13,2% |
| Muriithi et al. (13) | 47% | 38% | 14% |
| Tao Su et al. (26) | 45,9% | 22,9% | 29,2% |
| Yun et al. (71) | 41,1% | 10,8% | 45,1% |
| Hadded (37) | 64% | 10% | 26% |
| Our serie | 46,42% | 25% | 28,57% |
At the end of follow-up in our series, two of our patients were on end-stage hemodialysis. This result is comparable with those published in the literature, with a rate varying between 4% and 14% in the majority of studies [13, 23, 33, 47, 52].
In AIN, mortality is rarely related to renal damage per se [27]. Rather, it is the consequence of infectious complications and cardiovascular and neoplastic comorbidities. Indeed, the mortality rate in AIN described in the literature varies from 4% in Chabchoub's study to 17% in Clackson's and Yun's series [33, 36, 71]. There were no deaths in the acute phase of AIN in our study. The only death was of undetermined cause remote from the AIN episode.
Strengths and limitations of the study
The limitations of our study were primarily the number of cases and the retrospective nature of our research. However, in the literature, most studies on AKI were retrospective and based on data from renal biopsy registries.
The main strengths were the diversity of pathological conditions associated with AIN including some exceptional causes.