Effects of Nonsteroidal Anti-Inflammatory Drug Use on Early Development of Chronic Kidney Disease in Older Adults: A 10-Year Longitudinal Analysis Using the Korean Senior Cohort = 비스테로이드성 소염진통제(NSAIDs) 사용과 초기 만성신장질환 위험의 연관성: 한국 노인 코호트를 이용한 10년 추적 인구기반 연구

국문 초록 (Abstract)

배경: 비스테로이드성 소염진통제(Nonsteroidal Anti-inflammatory Drugs, NSAIDs)는 통증 및 염증 조절에 널리 사용되며, 특히 근골격계 질환이 흔한 노인에서 처방 빈도가 높다. 그러나 NSAIDs의 신독성(nephrotoxicity)에 대한 우려가 지속되고 있다. 고령 환자는 신기능 예비력이 감소하고 혈역학적 변화에 취약하여, 장기 복용 시 신장 손상의 위험이 증가할 가능성이 크다. 급성신손상(AKI)에 대한 영향은 잘 알려져 있으나, 장기간의 NSAID 사용이 초기 단계의 만성신장질환(CKD) 발생에 미치는 영향은 여전히 논란이 많다.
목적: 본 연구의 주된 목적은 기저 신기능이 정상인 지역사회 거주 노인을 대상으로 만성 NSAID 사용과 초기 CKD 발생 위험 간의 연관성을 규명하는 것이다. 또한 이차적으로 eGFR 저하율의 차이와 COX-1 억제제(우세) 사용자 군 및 COX-2 선택적 억제제 간 위험도의 차이를 비교하였다.
방법: 2009년부터 2019년까지 국민건강보험공단 노인코호트(NHIS-SC) 자료를 이용한 후향적 코호트 연구를 수행하였다. 연구대상은 65세 이상, 기저 eGFR 60 mL/min/1.73 m² 이상인 노인으로, NSAID 장기 복용군은 약물소유비율(Medication Possession Ratio, MPR) ≥80% 및 누적 처방기간 ≥1개월로 정의하였다. 연령, 성별, 동반질환(당뇨병, 고혈압, 고지혈증 등), Charlson 동반질환지수(CCI), 심혈관질환 이력 등을 보정하기 위해 1:2 성향점수매칭(PSM)을 실시하였다. 주요 결과변수는 추적기간 동안 eGFR이 60 mL/min/1.73 m² 미만으로 감소하고, 기저치 대비 ≥10% 감소를 보인 경우로 정의하였다.
결과: 최종 분석대상은 NSAID 사용자 604명과 비사용자 1,208명(총 1,812명)이었으며, 두 군의 기저 특성은 매칭 후 균형을 이뤘다. 10년 추적 결과, NSAID 장기 복용군은 CKD 발생 위험이 유의하게 높았다(HR 1.46, 95% CI 1.11–1.93). 또한 NSAID 사용군에서 eGFR 저하 속도가 유의하게 빨랐으며, COX-1 우세 억제제(HR 1.53)와 COX-2 선택적 억제제(HR 1.61) 간 위험도 차이는 통계적으로 유의하지 않았다. 이는 특정 약물군이 아닌 NSAID 계열 전체의 ‘class effect’를 시사한다. 말기신부전(ESRD) 발생은 드물었으며 군 간 유의한 차이는 없었다.
결론: 한국 노인 인구를 대상으로 한 본 대규모 인구기반 연구에서 NSAID의 장기 사용은 COX 선택성과 관계없이 초기 CKD 발생 위험 증가 및 신기능 저하 가속과 독립적으로 관련되어 있었다. 이에 따라 노인 환자에서 NSAID의 장기 처방 시 신중한 위험-이익 평가, 대체 통증 관리 전략의 우선 고려, 및 주기적 신기능 모니터링이 필요함을 강조한다.

다국어 초록 (Multilingual Abstract)

Background: Nonsteroidal anti-inflammatory drugs (NSAIDs) are ubiquitous in the management of pain and inflammation, particularly among older adults burdened by musculoskeletal conditions. While effective, their use is tempered by concerns regarding nephrotoxicity. The aging kidney, characterized by reduced functional reserve and increased susceptibility to insults, may be particularly vulnerable. Although the acute hemodynamic effects of NSAIDs, mediated by prostaglandin inhibition, are well-established causes of acute kidney injury (AKI), the long-term impact of chronic NSAID exposure on the initiation and progression of chronic kidney disease (CKD) in the general senior population remains contentious. Previous studies have yielded conflicting results, often limited by methodological issues such as confounding by indication, short follow-up durations, and inadequate assessment of medication adherence.
Objectives: This study aimed to rigorously investigate the association between chronic NSAID use and the risk of developing early-stage CKD in a large, representative cohort of community-dwelling older adults with initially preserved kidney function. Secondary objectives included evaluating the rate of estimated glomerular filtration rate (eGFR) decline and comparing the risks associated with Cyclooxygenase-1 (COX-1) predominant inhibitors versus COX-2 selective inhibitors.
Methods: This retrospective cohort study utilized data from the National Health Insurance Service-Senior Cohort (NHIS-SC) in South Korea (2009-2019). The study population comprised individuals aged 65 years or older with a baseline eGFR≥60 mL/min/1.73m2. Chronic NSAID exposure was stringently defined using a medication possession ratio (MPR)≥80% for a duration of at least one month. To minimize confounding, 604 NSAID users were matched 1:2 with 1,208 non-users using propensity score matching, balancing demographics, comorbidities (diabetes, hypertension, hyperlipidemia), Charlson Comorbidity Index (CCI), and cardiovascular history. The primary outcome was the development of kidney dysfunction, defined as eGFR<60 mL/min/1.73 m2 coupled with a ≥10% decline from baseline eGFR (calculated via the CKD-EPI equation). Hazard ratios (HRs) and 95% confidence intervals (CIs) for CKD were estimated using Cox proportional hazards regression models.
Results: The matched cohort (N=1812) was well-balanced across baseline characteristics. Over the 10-year follow-up period, chronic NSAID use was significantly associated with an increased risk of developing CKD (HR 1.46; 95% CI 1.11-1.93) compared to non-use. Longitudinal analysis confirmed a faster rate of eGFR decline among NSAID users. Subgroup analyses demonstrated elevated risks for both COX-1 inhibitor (predominant) user group (HR 1.53) and COX-2 selective inhibitors (HR 1.61), suggesting a potential class effect rather than a differential risk based on COX selectivity. The incidence of end-stage renal disease (ESRD) was rare in this early CKD cohort and did not significantly differ between groups.
Conclusion: In this large-scale, population-based study of older adults, chronic use of NSAIDs, regardless of COX selectivity, was independently associated with an increased risk of incident early CKD and accelerated kidney function decline. These findings underscore the necessity for judicious prescribing of NSAIDs in the senior population. Clinicians must carefully weigh the risks and benefits, prioritize alternative pain management strategies when feasible, and implement rigorous renal function monitoring during long-term NSAID therapy.

목차 (Table of Contents)

• Abstract i
• Chapter 1. Introduction 1
• 1.1 Background and Significance of the Study 1
• 1.2 Renal Physiology and NSAID Pharmacology 4
• 1.3 Mechanisms of NSAID-Induced Nephrotoxicity 8
• 1.4 Review of the Literature 11
• 1.5 Study Rationale and Objectives 15
• Chapter 2. Materials and Methods 18
• 2.1 Study Design and Data Source 18
• 2.2 Study Population and Selection/Exclusion Criteria 19
• 2.3 Exposure Assessment and Application of MPR 21
• 2.4 Outcome Assessment 22
• 2.5 Covariate Assessment 23
• 2.6 Statistical Analysis and Application of PSM 24
• 2.7 Ethical Considerations 26
• Chapter 3. Results 27
• 3.1 Study Participant Selection Process and Cohort Characteristics 27
• 3.2 NSAID Use and the Risk of CKD Development 31
• 3.3 Secondary Outcomes and Subgroup Analysis 33
• 3.4 Sensitivity Analysis Results 37
• Chapter 4. Discussion 38
• 4.1 Summary and In-depth Interpretation of Major Findings 38
• 4.2 Delineating the Scientific Position of This Study Through Comparison
• with Prior Research 39
• 4.3 Potential Mechanisms of Chronic Nephrotoxicity and Their Relevance
• to Study Findings 40
• 4.4 COX Selectivity and Renal Safety: An In-depth Examination
• of the Class Effect 43
• 4.5 Clinical Implications and Public Health Significance 47
• 4.6 Strengths and Limitations 51
• 4.7 Recommendations for Future Research 56
• Chapter 5. Conclusion 58
• List of Abbreviations 59
• Bibliography 60
• Supplementary 62
• Abstract in Korean 71