Retrospective clinical and pharmacological audit of non-steroidal anti-inflammatory drug prescriptions and the "Triple Whammy" phenomenon in polymorbid outpatients

Introduction

Non-steroidal anti-inflammatory drugs (NSAIDs) are, in most cases, the first line of treatment for pain of various etiologies [1, 2]. The widespread availability of over-the-counter formulations ensures their high accessibility [3, 4]. However, the use of this drug class is associated with the risk of adverse drug reactions (ADRs), primarily affecting the gastrointestinal tract, cardiovascular system, and kidneys [5]. In outpatient practice, NSAIDs are prescribed by physicians of various specialties, which often leads to insufficient consideration of systemic risk factors and drug–drug interactions [6–8].

A particular problem is presented by comorbid therapeutic patients suffering from arterial hypertension (AH), chronic heart failure (CHF) [5, 6, 9–11], and chronic kidney disease (CKD). The very prescription of NSAIDs may pose a risk of decompensation for patients with pre-existing CKD [12–15]. Such patients are long-term recipients of RAAS inhibitors (ACEIs or ARBs) and diuretics [16]. Adding systemic NSAIDs to this regimen leads to the formation of the "Triple Whammy" phenomenon [7, 17–19]. Combined inhibition of prostaglandins in the setting of angiotensin II blockade and diuresis-induced hypovolemia sharply reduces glomerular filtration pressure and is a leading predictor of acute kidney injury [18].

Despite the known nature of the "Triple Whammy," data on its prevalence in real-world outpatient practice remain limited. In outpatient settings, pharmacovigilance aimed at minimizing such risks becomes particularly important.

Objective

To evaluate the prescribing patterns of systemic NSAIDs and the frequency of potentially nephrotoxic drug combinations in polymorbid outpatients.

Study objectives

1. To analyze the structure of NSAID prescriptions and the quality of medical record documentation.

2. To evaluate the frequency of potentially dangerous drug interactions depending on baseline renal function.

3. To analyze the clinical structure of possible drug-associated events recorded in medical documentation, including lack of therapeutic effect.

Materials and Methods

Study design and sample. A retrospective single-center study was conducted based on a clinical and pharmacological audit of outpatient records of patients receiving NSAIDs. The data source was records from the outpatient electronic health records of the Samara Regional Medical Information and Analytical Center (MIAC SO) for the period from January 1, 2024, to December 31, 2024. All information was anonymized. The sample (n=50) was formed using consecutive (sequential) sampling of outpatient records of patients who successively sought medical care during the period from January 1, 2024, to December 31, 2024, and who met the established inclusion criteria. This approach ensured the absence of subjectivity in the formation of the study group and the reproducibility of the design in similar clinical settings.

Inclusion criteria:

1. Regular outpatient follow-up during the study period.

2. Prescription of systemic or topical NSAIDs.

3. Presence of concomitant polymorbid pathology (gastrointestinal diseases, cardiovascular diseases, risk factors for gastrointestinal bleeding).

4. Presence in the medical records of data on serum creatinine levels at the time of (or immediately before) initiation of NSAID therapy.

Exclusion criteria: history of oncological diseases; palliative status; death during the observation period; pregnancy and lactation. The final sample included data from 50 outpatient records.

Data collection and clinical assessment. The estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI formula [20].

Patients were stratified into two groups:

1. Preserved renal function (eGFR ≥60 mL/min/1.73 m²).

2. Clinically significant reduction (eGFR <60 mL/min/1.73 m², CKD stages 3–5).

The frequency of co-prescription of potentially nephrotoxic combinations ("Triple Whammy" = systemic NSAIDs + RAAS inhibitors + diuretic) was analyzed. Only cases of systemic NSAID prescription were considered when calculating the prevalence of this phenomenon.

Documentation deficiency was defined as the absence of indication of the maximum duration of analgesic therapy and/or the maximum daily dose.

ADRs in this study were defined as clinical events recorded in medical documentation and retrospectively classified according to WHO-UMC [21] as possible/probable, as well as lack of therapeutic effect when a particular drug was prescribed. Causality assessment was performed by consensus. For the probable category, a strict temporal relationship and positive improvement after drug withdrawal were required, in the absence of more likely alternative causes.

This study was a retrospective clinical and pharmacological audit. In accordance with the principles of the Declaration of Helsinki, working with fully anonymized data obtained from the MIAC SO electronic health records does not require ethics committee approval or informed consent, since the study does not involve intervention in the treatment process, and patient identity cannot be identified.

Statistical analysis. Statistical analysis was performed using Python 3 with libraries (pandas, scipy.stats). Non-parametric methods were applied: quantitative variables are presented as median (Me) and interquartile range (IQR). Categorical variables are presented as absolute numbers and proportions (n, %). Comparisons of binary feature frequencies were performed using Fisher's exact test (two-tailed). To estimate the strength of association, the odds ratio (OR) with 95% confidence interval (95% CI) was calculated. Differences were considered statistically significant at p <0.05. Despite the wide confidence interval, the lower bound of the 95% CI (>1) confirms a statistically significant association and the clinical relevance of the observed effect.

Additionally, a post-hoc power calculation was performed for the primary comparison of the frequency of the "Triple Whammy" pattern between groups (eGFR ≥60 and <60 mL/min/1.73 m²) at α=0.05 (two-tailed). The calculation was conducted for comparison of two proportions using Cohen's h effect size. With observed proportions of 12.5% (3/24) and 61.5% (16/26), the post-hoc power was approximately 0.97, indicating a high probability of detecting an effect of this magnitude in the current sample. It should be noted that post-hoc power depends on the observed effect and does not replace interpretation of the 95% CI.

Results

Within the outpatient audit:

1. The frequency of the "Triple Whammy" pattern in outpatient settings was quantitatively estimated (38% of cases).

2. A statistically significant association was demonstrated between the most nephrotoxic drug combination and the presence of pre-existing reduced renal function (eGFR <60 mL/min/1.73 m²).

3. A high level of systemic documentation deficiencies was recorded (absence of therapy limits in 64% of cases), which may contribute to cascading polypharmacy.

4. The absolute risk of forming the "Triple Whammy" pattern in patients with reduced renal function was 61.5% versus 12.5% in those with preserved eGFR (risk difference 49.0%). The NNH (Number Needed to Harm) ≈ 2, which may reflect a high absolute risk difference between groups (Figure 1), and requires further investigation.

5. Among prescribed systemic NSAIDs, nimesulide (16%), ketorolac (18%), aceclofenac (14%), and meloxicam (12%) predominated; the proportion of other drugs did not exceed 10% in each case.

Figure 1. Frequency of formation of the potentially nephrotoxic combination "Triple Whammy" depending on the initial level of glomerular filtration rate (p <0.001)

Baseline characteristics of the sample (Figure 2) reflect a typical profile of a comorbid therapeutic patient: 23 males and 27 females, median age 55.5 years (IQR: 41.0–69.0), median BMI 25.6 kg/m² (IQR: 23.4–27.1). Among indications for NSAID prescription, osteochondrosis/dorsopathy (56%, n=28) and osteoarthritis (42%, n=21) predominated.

Figure 2. Structure of indications for the prescription of systemic NSAIDs (n=50; more than one indication was allowed)

The audit of medical record quality revealed clinical monitoring deficiencies: in 64% of cases (n=32), the maximum duration of NSAID use was not specified (prescriptions "as needed for pain"), which is a risk factor for uncontrolled use. In 74% (n=37) of the analyzed cases, subsequent pharmacotherapy adjustment was required (Figure 3).

Figure 3. Need for adjustment of pharmacotherapy and structure of interventions (n=50; >1 intervention was allowed)

Signs of CKD stages 3–5 (eGFR <60 mL/min/1.73 m²) were present in more than half of patients—52% (n=26). The median eGFR in the preserved function group was 85.6 mL/min/1.73 m² (IQR: 80.2–109.5), and in the CKD group—47.3 mL/min/1.73 m² (IQR: 42.4–52.7). Data analysis (Table 1) revealed that the frequency of the "Triple Whammy" pattern was 38% overall (19 of 50 cases) (Figure 3). This combination was statistically significantly more frequent in the group of patients with pre-existing eGFR reduction (p <0.001). Reduced eGFR was associated with an increased likelihood of prescribing the "Triple Whammy" combination (OR 11.2 [95% CI: 2.64–47.48]). The observed wide confidence interval is due to the limited sample size, which determines high variance of the point estimate. The absolute risk difference for the "Triple Whammy" pattern between groups was 49.0%. The relative risk of prescribing this combination in patients with reduced eGFR was 4.92. The calculated NNH ≈ 2 indicates that for every two patients with reduced renal function, one additional case of a potentially nephrotoxic combination occurs.

Table 1

Frequency of potentially nephrotoxic co-prescriptions and development of drug-associated events, including lack of treatment effect, depending on the level of renal function (N=50)

In the overall sample, ADRs (possible/probable by WHO-UMC), as well as lack of therapeutic effect, were recorded in 14 cases (28%). Detailed analysis allowed identification of the following predominant complication patterns (Figure 4):

1. Cardiovascular events (n=7). The most common manifestation was blood pressure destabilization, up to severe hypertensive crises. Complications developed early—on days 2–7 after starting treatment. In four cases, the severity required emergency hospitalization; in the remaining cases, emergency medical service calls were recorded. After drug withdrawal, clinical stabilization was noted.

2. Renal impairment (n=3). Clinically significant increases in serum creatinine were recorded. In one case, a critical rise in creatinine from baseline 218 to 280 µmol/L was documented on day 8 during intramuscular administration of ketorolac (requiring hospitalization).

3. Hemorrhagic and neurological complications. Isolated cases of severe events were identified, including uterine bleeding (hospitalization on day 8 of ketorolac intake), as well as the formation of probable drug-induced (analgesic-overuse) headache.

Figure 4. Frequency of potentially dangerous co-prescriptions (n=50)

Separately, lack of therapeutic effect (n=2) should be noted: persistence of severe pain syndrome, likely due to inappropriate drug or dose selection, which required revision of the analgesic regimen. In the subgroup of patients receiving ketorolac (n=9), ADRs were recorded in more than half of cases (n=5), which is a significant signal requiring further investigation.

Figure 5. Structure of registered drug-associated events against the background of NSAIDs, including lack of effect (n=14)

Discussion

This clinical and pharmacological audit demonstrated that a significant proportion of systemic NSAID prescriptions in outpatient practice are associated with deficiencies in documenting safety parameters. The absence of documented therapy limits ("as needed for pain") in 64% of records complicates monitoring and correlates with known risks of severe cardiovascular and renal complications with prolonged analgesic use.

Our findings illustrate a classic example of a situation against which the 2024 KDIGO guidelines are directed: patients with CKD stages 3–5 continue to receive NSAIDs in combination with RAAS blockers and diuretics without systematic monitoring, highlighting the gap between clinical guidelines and real-world clinical practice.

The analysis showed a high overall frequency of the "Triple Whammy" combination (38%) (95% CI: 25–52%). The identified pattern—a multiple predominance of this combination (OR 11.2) in patients with pre-existing reduced eGFR—is likely related to the comorbidity structure and the need for RAAS inhibitors and diuretics in patients with reduced renal function. The prescription of systemic NSAIDs in these conditions increases the likelihood of a potentially nephrotoxic cascade. This may indicate insufficient oversight by treating physicians: analgesics are prescribed in isolation for pain relief without due consideration of drug–drug interaction risks. The wide confidence interval (2.64–47.48) reflects the small sample size, thus the exact effect size requires confirmation in larger studies.

The frequency of the "Triple Whammy" combination in our sample (38% among patients receiving systemic NSAIDs) is in the upper range described in pharmacoepidemiological studies and clinical pharmacological audits [22] that assessed the prevalence of this phenomenon. At the same time, most published studies have focused either on hospitalizations for acute kidney injury or on inpatient practice, whereas our data reflect the outpatient stage of care, which may explain the higher proportion of potentially dangerous combinations.

Among registered drug-associated events, hypertensive crises predominated, which is pathogenetically explained by inhibition of prostacyclin synthesis and sodium retention, counteracting the effect of baseline antihypertensive drugs. The timing of renal impairment development (on day 8) is fully consistent with the literature on early (within the first 1–2 weeks) occurrence of acute kidney injury in the setting of the "Triple Whammy" pattern.

Of note, the most nephrotoxic combination more frequently occurs precisely in patients with pre-existing chronic kidney disease. This suggests that in outpatient practice, decisions regarding analgesic prescriptions are made without comprehensive assessment of drug–drug interactions and cumulative nephrological risk. Thus, the problem is not individual but systemic, reflecting a deficit of integrated clinical decision support mechanisms.

Although the small sample size resulted in a wide confidence interval for the odds ratio, the final indicators indicate a pronounced clinical threat. The absolute risk difference of 50% and an NNH of 2 suggest an extremely high probability of the "Triple Whammy" pattern occurring in every second patient with CKD. This underscores the need to move from isolated relative risk assessment to baseline patient profiling, which is the gold standard in modern NSAID safety research [23].

To optimize outpatient practice and minimize risks, we propose the implementation of a standardized five-step safety algorithm (checklist):

1. Assessment of renal function: mandatory eGFR calculation before initiating NSAID therapy.

2. Medication list screening: checking for RAAS inhibitors and diuretics (avoidance of the "Triple Whammy" cascade).

3. Limiting: selection of the minimum effective NSAID dose for the shortest duration, with clear documentation in the record.

4. Monitoring: creatinine and blood pressure (BP) control at 48–72 hours in high-risk patients (elderly, CKD stages 3–5, CHF, hypovolemia).

5. Alternatives: priority consideration of topical NSAIDs or paracetamol (taking into account hepatotoxicity risks with chronic use and total dose control).

Publications on the implementation of such clinical decision support systems demonstrate high algorithm sensitivity, acceptable specificity, and a substantial proportion of clinician-accepted interventions, leading to a reduction in the frequency of dangerous prescriptions and enabling control over potentially dangerous drug combinations in routine practice without significantly increasing the physician's workload [24–26].

Study Limitations

This study has limitations due to its pilot, single-center, retrospective design and small sample size (n=50), which limits statistical power and precludes causal inferences. Selection bias is possible, since the inclusion criterion was the availability of creatinine data. The analysis of drug-associated events is based on events recorded in outpatient documentation and may underestimate their true frequency; most cases were classified as "Possible" according to WHO-UMC. Due to the limited number of events, multivariable analysis was not performed, which does not exclude the influence of confounding factors. Nevertheless, the magnitude of the absolute risk difference and the low NNH value may indicate the clinical significance of the identified associations, warranting further research.

Conclusions and Practical Significance

1. In outpatient practice, a high level of documentation deficiencies was identified: in 64% (n=32) of cases, data on the maximum duration of NSAID therapy are absent, necessitating the implementation of a standardized checklist for internists.

2. Prescription of the potentially nephrotoxic drug combination "Triple Whammy" (systemic NSAIDs + RAAS inhibitors + diuretic) is statistically significantly associated with reduced renal function (OR 11.2 [95% CI: 2.64–47.48]). The observed trend requires confirmation of the exact effect size in larger studies.

3. In the clinical structure of registered complications during systemic NSAID therapy (including ketorolac), blood pressure destabilization and worsening renal function predominate.

4. The results obtained indicate the need to implement algorithms for automated risk alert at the level of regional medical information systems, and also require confirmation of the exact effect size in larger and, preferably, multicenter studies.