Postoperative Pain in Patients Receiving Ketoprofen After Total Hip Arthroplasty: The Role of Pharmacogenetics

Abstract

Background: Ketoprofen is one of the most commonly prescribed NSAIDs; however, its pharmacogenetics remains poorly understood. The objective was to evaluate the influence of patients’ pharmacogenetic profiles on the effectiveness of ketoprofen for postoperative pain management after total hip arthroplasty, including postoperative analgesia (pain levels, opioid consumption) and the incidence of adverse reactions during hospitalization and up to 12 months post-surgery. Methods: The study included 53 patients (31 (58.49%) women, median age 66.0 [60.0–74.0] years) undergoing total hip arthroplasty. Genotyping was performed using real-time PCR to analyze 18 single-nucleotide polymorphisms (SNPs) across the following genes: CYP2C9 (rs1799853, rs1057910), CYP2C8 (rs10509681, rs11572080), CYP3A4 (rs35599367), CYP3A5 (rs776746), UGT2B7 (rs73823859, rs7439366, rs7668282), ABCB1 (rs1045642, rs4148738, rs2032582, rs1128503), PTGS1 (rs10306135, rs12353214), PTGS2 (rs20417), C3orf20 (rs12496846), and ZNF493-ZNF429 (rs2562456). Results: We did not find significant associations between patients’ genotypes and pain levels or postoperative opioid analgesic consumption or adverse reactions when ketoprofen was used for pain management in patients undergoing total hip arthroplasty. Conclusions: Routine pharmacogenetic testing for ketoprofen is not supported by our findings.

1. Introduction

Ketoprofen is a racemic mixture of S(+)- and R(-)-enantiomers, with only the S(+)-component, dexketoprofen, exhibiting pharmacological activity. Ketoprofen is primarily metabolized by phase II biotransformation enzymes, specifically UDP-glucuronosyltransferases, through glucuronidation. The main enzymes involved are UGT2B7, UGT2B4, UGT1A3, and UGT1A9 [1,2,3]. The cytochrome P450 system, particularly CYP2C9, is believed to play a lesser role in ketoprofen metabolism [3,4,5,6]. The approximate contribution of glucuronidation by UGT enzymes compared to oxidation by CYP2C9 is estimated to be about 80% to 20%, as up to 80% of the ketoprofen dose is excreted as acyl glucuronide [7]. The role of drug transporters in ketoprofen pharmacokinetics remains unclear, as no data indicate whether ketoprofen is a substrate for solute carrier (SLC) membrane transporters or P-glycoprotein (encoded by ABCB1, ATP-binding cassette subfamily B member 1) [8].

Results from genome-wide association studies (GWASs) suggest that certain genetic factors may influence individual differences in analgesic sensitivity or opioid analgesic requirements. The C3orf20 gene encodes the chromosome 3 open reading frame 20 protein, whose characteristics and functions remain unknown. It has been found that the rs12496846 single-nucleotide polymorphism (SNP) of C3orf20 may predict analgesic requirements; specifically, the G allele was possibly associated with lower opioid sensitivity and, consequently, greater opioid analgesic requirements following painful cosmetic orthognathic surgery and major open abdominal surgery [9]. Another SNP, rs2562456, located in the unknown gene LOC400680, showed a significant association with the onset time of analgesia after ketorolac administration [10]. The gene encoding zinc finger protein 429 (ZNF429) was found to be in linkage disequilibrium with this SNP. Zinc finger proteins regulate the processes that modulate the frequency, rate, or extent of DNA-dependent transcription by binding to DNA. However, the specific biological function of ZNF429 in analgesic drug metabolism and pain pathways has not yet been reported. Although the results of these GWASs were not directly applicable to this study, these SNPs were included in the genotyping panel to clarify their role in postoperative pain management in patients undergoing total hip arthroplasty treated with ketoprofen.

The Clinical Pharmacogenetics Implementation Consortium guidelines (2020) do not mention ketoprofen or dexketoprofen as either CYP2C9-dependent nonsteroidal anti-inflammatory drugs or as drugs with alternative metabolic pathways [11]. It is likely that there are currently insufficient data on the pharmacogenetic characteristics of patients in relation to their response to these drugs to draw definitive conclusions or make therapeutic recommendations. Therefore, further pharmacogenetic studies on ketoprofen remain necessary.

In the Russian Federation, ketoprofen ranked sixth among nonsteroidal anti-inflammatory drugs (NSAIDs) in sales volume in both retail and commercial markets, with 260,339,866 packages sold over a 10-year period from 2010 to 2020 [12]. During the same 10-year period from 2010 to 2020, there were 964 reports in the national database “Pharmacovigilance” (automated information system of Roszdravnadzor, Russian Federal Service for Surveillance in Healthcare), placing ketoprofen seventh among NSAIDs in terms of the number of reports, accounting for 4.8% of the total spontaneous reports for all NSAIDs, which amounted to 20,088. Additionally, ketoprofen was among the most prescribed NSAIDs in hospitals across Russia [12].

The objective was to evaluate the impact of patients’ pharmacogenetic features on the effectiveness of ketoprofen for postoperative pain management after total hip arthroplasty. The assessment included the effectiveness of postoperative analgesia, measured by pain levels and opioid analgesic consumption, as well as the occurrence of adverse reactions during hospitalization and within 12 months after surgery, including chronic pain and long-term adverse effects.

2. Materials and Methods

2.1. Patients and Clinical Outcomes

This open, prospective observational study was conducted over a period of 1 year and 9 months, from February 2024 to November 2025. It included an inpatient phase (February 2024 to October 2024) and follow-up assessments at 2, 6, and 12 months post-surgery.

This was a real-world study conducted without investigator intervention in prescribing practices or routine clinical processes, except for the introduction of a patient pain assessment diary and blood sample collection.

The study was approved by the Local Ethics Committee of N.I. Pirogov City Clinical Hospital No. 1 (protocol No. 11 dated 21 December 2023). All participants provided written informed consent for participation in the study, as well as for the collection and storage of DNA samples.

The study included 53 patients receiving inpatient treatment in the Trauma Department of N.I. Pirogov City Clinical Hospital No. 1, Moscow.

Inclusion criteria for the study were age over 18 years, hospitalization for total hip arthroplasty, and written informed consent to participate. Exclusion criteria included contraindications to the use of non-steroidal anti-inflammatory drugs (NSAIDs), psychiatric disorders that could hinder adherence to the study protocol, severe comorbid conditions, inability to undergo follow-up and monitoring, previous inflammatory processes in the joint, and the need for revision surgeries.

The investigator collected detailed data on NSAID use during the outpatient phase, comorbidities, and concomitant therapies. The effectiveness of pain relief was assessed by postoperative pain levels and the need for additional analgesia with opioid medications, converting all opioid dosages into morphine milligram equivalents (MME). Safety was evaluated based on laboratory parameters—including creatinine clearance, urinalysis, complete blood count, and transaminase levels—measured upon admission and discharge, as well as documented cases of adverse reactions and complications during hospitalization. After discharge, patients were surveyed by phone at 2, 6, and 12 months post-surgery to assess pain levels at the surgical site, analgesic use, requests for medical assistance, and any adverse reactions following discharge.

The Naranjo Adverse Drug Reaction Probability Scale, developed to assess the relationship between an adverse event and drug administration and to classify it as an adverse drug reaction, was used [13]. Only “definite” and “probable” adverse reactions were included in the study.

2.2. Pain Assessment

To assess pain intensity during hospitalization, the Numerical Rating Scale (NRS) was used—an instrument that allows patients to self-rate their pain severity on a scale from 0 (no pain) to 10 (unbearable pain).

Patients completed a paper pain assessment diary from the time of hospitalization until discharge, recording their pain levels at least three times daily—at rest and during movement—at fixed times: 8:00 AM, 2:00 PM, and 8:00 PM. They also had the option to record additional pain assessments outside these intervals if their pain intensified. Based on the date and time of surgery, researchers categorized the pain scores into at least 11 periods: before surgery; up to 6 h post-operation (p/o); 7–18 h p/o; 19–30 h p/o; 31–42 h p/o; 43–48 h p/o; 49–66 h p/o; 67–78 h p/o; 79–90 h p/o; 91–102 h p/o; and 103–114 h p/o. Some time intervals were combined to unify the data due to the absence of pain assessments during evening and night hours, specifically from 8:00 PM to 8:00 AM.

At two, six, and twelve months after surgery, patients were surveyed by phone regarding their current pain levels at the surgical site, analgesic use, and any requests for medical assistance during the preceding period.

2.3. Postoperative Pain Management with Ketoprofen

All 53 patients received a single intravenous injection of 100 mg ketoprofen within the first hour after surgery, followed by an assessment of its analgesic efficacy 30 min later.

Monotherapy with ketoprofen at a dosage of 200 mg/day was effective in 2 patients (3.8%) on the first day post-surgery. Additional analgesics from other classes were required by 51 patients (96.2%).

Among them, 8 patients (15.1%) with mild pain syndrome (NRS scores ≤ 4), were additionally prescribed paracetamol, which was effective in only one case (1.9%). The remaining patients required tramadol (n = 3, 5.7%) due to insufficient analgesic effect of ketoprofen combined with paracetamol and increasing pain intensity to 4–7 points on the NRS; morphine (n = 3, 5.7%) for pain levels exceeding 7 points on the NRS; and sequentially tramadol and morphine due to insufficient pain relief (n = 1, 1.9%).

Patients with NRS scores of 4–7 were prescribed tramadol (n = 20, 37.7%) following the initial injection of ketoprofen. In 13 cases (24.5%), the analgesic effect was insufficient, requiring an additional morphine injection. Patients with an NRS score greater than 7 were prescribed morphine (n = 22, 41.4%) or trimeperidine (n = 1, 1.9%), a Russian analog of meperidine (

Table 1. Postoperative Pain Management During the First Day After Surgery in Patients Undergoing Total Hip Arthroplasty.

Patients (n, %)	Drugs During the First Day Post-Surgery
	Ketoprofen 100 mg Intravenously	Paracetamol 1000 mg Intravenously	Tramadol 100 mg Intramuscularly	Morphine 10 mg Intramuscularly	Trimeperidine * 20 mg Intramuscularly
22 (41.4)	+	−	−	+	−
13 (24.5)	+	−	+	+	−
7 (13.2)	+	−	+	−	−
3 (5.7)	+	+	+	−	−
3 (5.7)	+	+	−	+	−
2 (3.8)	+	−	−	−	−
1 (1.9)	+	+	+	+	−
1 (1.9)	+	−	−	−	+
1 (1.9)	+	+	−	−	−

* Trimeperidine, or 1,2,5-trimethyl-4-propionyloxy-4-phenylpiperidine hydrochloride, is a Russian opioid analgesic and an analog of meperidine [14].

).

Starting on the second day after surgery, all patients received standard postoperative analgesia with ketoprofen at a dose of 100 mg administered intravenously twice daily.

The study was conducted without the investigators intervening in the prescription of analgesic therapy during the postoperative period.

2.4. Opioid Analgesic Consumption

The total amount of opioid analgesics prescribed during hospitalization was summed and converted into morphine milligram equivalents (MME), based on the conversion ratios in which 100 mg of tramadol or 20 mg of trimeperidine are equivalent to 10 mg of injectable morphine [15]. This serves as a universal metric for quantifying opioid analgesic consumption. The choice of opioid analgesic was made independently, without investigator intervention.

2.5. Selection of Candidate Genes and Molecular Genetic Study

Blood samples for pharmacogenetic studies were collected from 51 patients. Due to technical issues, blood samples could not be obtained from two patients. Pharmacogenetic data were analyzed retrospectively after the completion of treatment.

Venous blood was drawn from the patients’ antecubital veins into vacuum tubes (Greiner Bio-One, Kremsmünster, Austria) containing K2-EDTA. Blood samples were stored at −80 °C until analysis. Genomic DNA was extracted from whole blood using a DNA extraction kit with a sorbent (Syntol, Moscow, Russia). The quantity and quality of the extracted DNA were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Genotyping was performed using real-time PCR on a CFX96 Touch Real-Time System with CFX Manager software version 3 (Bio-Rad, Hercules, CA, USA), utilizing commercial reagent kits (Syntol, Moscow, Russia; TestGen, Ulyanovsk, Russia). A total of 18 single-nucleotide polymorphisms (SNPs) in the following genes were analyzed in patients: CYP2C9 (rs1799853, rs1057910), CYP2C8 (rs10509681, rs11572080), CYP3A4 (rs35599367), CYP3A5 (rs776746), UGT2B7 (rs73823859, rs7439366, rs7668282), ABCB1 (rs1045642, rs4148738, rs2032582, rs1128503), PTGS1 (rs10306135, rs12353214), PTGS2 (rs20417), C3orf20 (rs12496846), and ZNF493-ZNF429 (rs2562456). Syntol Assay ID: rs1799853—NP-456-100, rs1057910—NP-457-100, rs35599367—NP-725-100, rs776746—NP-467-100, rs1045642—NP-447-100, rs2032582—NP-497-100, rs1128503—NP-481-100.

The selection of pharmacogenetic markers was based on the following sources: the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines [11]—the study included candidate genes mentioned in the recommendations that may be associated with the pharmacokinetics and pharmacodynamics of some NSAIDs—namely, CYP2C9 (rs1799853, rs1057910), CYP2C8 (rs10509681, rs11572080), CYP3A4 (rs35599367), CYP3A5 (rs776746), PTGS1 (rs10306135, rs12353214), and PTGS2 (rs20417); and individual studies: UGT2B7 (rs73823859, rs7439366, rs7668282), due to ketoprofen metabolism through glucuronidation via UDP-glucuronosyltransferases [1,2,3]; ABCB1 (rs1045642, rs4148738, rs2032582, rs1128503), due to the possible role of P-glycoprotein in ketoprofen transport [8]; C3orf20 (rs12496846), based on data suggesting that this genetic polymorphism may predict analgesic requirements according to a genome-wide association study [9]; and ZNF493-ZNF429 (rs2562456), supported by genome-wide association study data indicating that this marker is associated with the onset time of analgesia when using NSAIDs, specifically ketorolac [10].

2.6. Statistical Analysis

Statistical analysis was performed using the StatSoft Statistica 12.0 software package. To assess the normality of the distribution of quantitative data, frequency histograms, the Kolmogorov–Smirnov test, and the Shapiro–Wilk test were used. For quantitative data with a normal distribution, the mean (M) and standard deviation (±SD) were calculated, and Student’s t-test was applied to analyze intergroup differences. Quantitative variables with non-normal distributions were expressed as the median (Me) and percentiles. Qualitative variables were presented as absolute values (n) and percentages (%).

No preliminary sample size calculation was performed. Instead, the sample size was determined based on the number of patients included in previous studies on postoperative pain management and pharmacogenetics, as well as the capabilities of the clinical center.

The genotype frequency distribution of the pharmacogenetic markers studied was tested for conformity with the Hardy–Weinberg equilibrium. All patients were divided into subgroups based on their genotypes: carriers of the polymorphic variant (heterozygotes and homozygotes) and “wild-type” homozygotes. Subsequently, these genetically defined subgroups were compared to identify associations with the patients’ clinical parameters. For the comparison of quantitative data, the nonparametric Mann–Whitney U test was applied. To assess differences between groups in qualitative variables, the χ2 test with Yates’ correction was used; when test assumptions were not met, Fisher’s exact two-tailed test was applied. To assess the relationship between the studied factors, the odds ratio (OR) with a 95% confidence interval (95% CI) was calculated.

Multiple comparisons were adjusted using the Bonferroni correction. Since no statistically significant differences were found, subsequent analyses were conducted using uncorrected p-values to allow readers to assess the statistical significance of the differences. Therefore, the findings should be considered preliminary and interpreted with caution pending external validation. A p-value < 0.05 was considered statistically significant.

3. Results

3.1. Clinical Data of Patients

A total of 53 patients were enrolled, including 22 men (41.51%) and 31 women (58.49%), with a median age of 66.0 [60.0–74.0] years. Patient data are summarized in

Table 2. Data of Patients Undergoing Total Hip Arthroplasty.

Indicator	Result
Age, years	66.0 [60.0–74.0]
Female gender	31 (58.49%)
Body mass index, kg/m2	28.5 [25.6–31.9]
Smoking	8 (15.1%)
NSAID use before hospitalization	50 (94.3%)
Duration of NSAID use before hospitalization:
>1 year	40 (75.5%)
6 months–1 year	9 (17.0%)
<3 months	1 (1.9%)
0	3 (5.6%)

.

3.2. Genotyping Results

The genotyping results for 51 patients are presented in

Table 3. Genotyping Results for Patients Undergoing Total Hip Arthroplasty (n = 51).

Gene	SNP	Genotype	n (%)	Allele Frequency (%)		Hardy–Weinberg Equilibrium
						χ2	p-Value
CYP2C9	rs1799853	CC	41 (80.4)	C (89.2)	T (10.8)	0.35	0.55
		CT	9 (17.7)
		TT	1 (1.9)
CYP2C9	rs1057910	CC	46 (90.2)	C (95.1)	A (4.9)	0.14	0.71
		AC	5 (9.8)
CYP2C8	rs10509681	TT	44 (86.3)	T (92.2)	C (7.8)	1.77	0.18
		TC	6 (11.8)
		CC	1 (1.9)
CYP2C8	rs11572080	CC	44 (86.3)	C (92.2)	T (7.8)	1.77	0.18
		CT	6 (11.8)
		TT	1 (1.9)
CYP3A4	rs35599367	CC	48 (94.1)	C (97.1)	T (2.9)	0.05	0.83
		CT	3 (5.9)
CYP3A5	rs776746	GG	45 (88.2)	G (93.1)	A (6.9)	2.77	0.10
		AG	5 (9.8)
		AA	1 (2.0)
UGT2B7	rs73823859	GG	50 (98.04)	G (99.0)	A (1.0)	0.01	0.94
		GA	1 (1.96)
UGT2B7	rs7439366	CC	16 (31.4)	C (55.9)	T (44.1)	0.001	0.97
		CT	25 (49.0)
		TT	10 (19.6)
UGT2B7	rs7668282	TT	47 (92.2)	T (96.1)	C (3.9)	0.08	0.77
		TC	4 (7.8)
PTGS1	rs10306135	AA	33 (64.7)	A (80.4)	T (19.6)	0.001	0.97
		AT	16 (31.4)
		TT	2 (3.9)
PTGS1	rs12353214	CC	36 (70.6)	C (84.3)	T (15.7)	0.07	0.79
		CT	14 (27.5)
		TT	1 (1.9)
PTGS2	rs20417	CC	40 (78.4)	C (87.3)	G (12.7)	2.18	0.14
		CG	9 (17.7)
		GG	2 (3.9)
ABCB1	rs1045642	CC	7 (13.7)	T (60.8)	C (39.2)	0.25	0.62
		CT	26 (51.0)
		TT	18 (35.3)
ABCB1	rs4148738	CC	12 (23.5)	C (51.0)	T (49.0)	0.49	0.48
		CT	28 (54.9)
		TT	11 (21.6)
ABCB1	rs2032582	AA	13 (25.5)	A (51.0)	C (49.0)	0.02	0.87
		AC	26 (51.0)
		CC	12 (23.5)
ABCB1	rs1128503	AA	12 (23.5)	A (51.0)	G (49.0)	0.49	0.48
		AG	28 (54.9)
		GG	11 (21.6)
C3orf20	rs12496846	AA	23 (45.1)	A (66.7)	G (33.3)	0.04	0.83
		AG	22 (43.1)
		GG	6 (11.8)
ZNF493-ZNF429	rs2562456	TT	37 (72.5)	T (83.3)	C (16.7)	2.55	0.11
		CT	11 (21.6)
		CC	3 (5.9)

UGT—UDP-glucuronosyltransferase; PTGS—Prostaglandin-endoperoxide synthase, Cyclooxygenase; ABCB1—ATP (adenosine triphosphate) binding cassette subfamily B member 1; C3orf20—Chromosome 3 Open Reading Frame 20; ZNF—Zinc finger protein.

.

The distribution of alleles and genotypes conformed to the expectations of the Hardy–Weinberg equilibrium, with p > 0.05 for all genetic variants.

According to the CPIC guidelines, patients were further subdivided into three groups based on their CYP2C9 phenotype (

Table 4. CYP2C9 Diplotypes and Predicted Phenotypes of Patients Undergoing Total Hip Arthroplasty.

Diplotype	Activity Score (AS)	n (%)	Predicted CYP2C9 Phenotype
CYP2C9*1/*1	2	36 (70.6)	Normal metabolizers
CYP2C9*1/*2	1.5	9 (17.6)	Intermediate metabolizers (AS = 1.5)
CYP2C9*1/*3	1	5 (9.8)	Intermediate metabolizers (AS = 1)
CYP2C9*2/*2		1 (2.0)
Total		51 (100)

).

3.3. Pain Level and Opioid Analgesic Consumption

We studied the association between postoperative pain intensity and genetic polymorphisms and found that patients with the TC + CC genotypes of rs10509681 CYP2C8 experienced significantly more pronounced pain at 43–48 h after surgery, both at rest and during movement, as well as at 49–66 h after surgery at rest, compared with carriers of the TT genotype, p < 0.05 (

Table 5. Intensity of Postoperative Pain, Opioid Analgesic Consumption, and CYP2C8 rs10509681 Genotype.

Outcome	CYP2C8 rs10509681 Genotype		p
	TT	TC + CC
Pain (NRS, before surgery, at rest)	2.48 ± 0.73	2.71 ± 0.49	0.41
Pain (NRS, before surgery, movement)	3.57 ± 0.66	4.00 ± 0	0.09
Pain (NRS, peak 1–30 h p/o, at rest)	7.09 ± 1.05	7.29 ± 0.76	0.64
Pain (NRS, peak 1–30 h p/o, movement)	8.61 ± 0.92	8.71 ± 0.76	0.79
Pain (NRS, 31–42 h p/o, at rest)	3.57 ± 0.66	3.86 ± 0.69	0.29
Pain (NRS, 31–42 h p/o, movement)	4.77 ± 0.83	5.29 ± 1.38	0.18
Pain (NRS, 43–48 h p/o, at rest)	3.34 ± 0.53	4.00 ± 1.41	0.02 *
Pain (NRS, 43–48 h p/o, movement)	4.34 ± 0.53	5.00 ± 1.41	0.02 *
Pain (NRS, 49–66 h p/o, at rest	3.09 ± 0.29	3.43 ± 0.53	0.02 *
Pain (NRS, 49–66 h p/o, movement)	4.16 ± 0.37	4.43 ± 0.53	0.10
Overall trend (all time points 0–114 h)	No consistent difference	No consistent difference	-
Opioid consumption (0–114 h p/o, MME)	12.27 ± 5.65	14.29 ± 5.35	0.38

Note: Pain scores were assessed at multiple postoperative time points using separate comparisons. The few statistically significant differences were not consistent across time points and should be interpreted as exploratory findings. Abbreviations: NRS, Numerical Rating Scale; p/o, post operation; MME, morphine milligram equivalents. * Significant at p < 0.05 (uncorrected).

).

We found that patients in both genotype groups experienced mild (NRS score ≤ 3) or moderate (NRS score 4–6) pain. Postoperative opioid analgesic consumption was comparable between the groups.

The groups of patients with TT and TC + CC genotypes of the rs10509681 CYP2C8 variant were comparable in clinical, demographic, and laboratory parameters.

No significant differences in pain levels or opioid analgesic requirements were observed among patient groups based on the other genotypes studied, including CYP2C9 (rs1799853, rs1057910), CYP2C8 (rs11572080), CYP3A4 (rs35599367), CYP3A5 (rs776746), UGT2B7 (rs73823859, rs7439366, rs7668282), ABCB1 (rs1045642, rs4148738, rs2032582, rs1128503), PTGS1 (rs10306135, rs12353214), PTGS2 (rs20417), C3orf20 (rs12496846), and ZNF493-ZNF429 (rs2562456), or CYP2C9 phenotype (normal metabolizers + intermediate metabolizers (AS = 1.5) vs. intermediate metabolizers (AS = 1)).

At two, six, and twelve months after surgery, patients across all genotypes and CYP2C9 phenotypes reported comparable pain levels.

3.4. Acute Kidney Injury

Two patients (3.7%) developed acute kidney injury during hospitalization, defined by an increase in serum creatinine of 0.3 mg/dL (26.4 µmol/L) within 2 days, according to the KDIGO 2012 criteria (

Table 6. Patients with Acute Kidney Injury in the Postoperative Period.

Patients	Baseline Creatinine (μmol/L)	Baseline Glomerular Filtration Rate CKD-EPI (mL/min/1.73 m2)	Creatinine over 2 Days (μmol/L)	Glomerular Filtration Rate CKD-EPI over 2 Days (mL/min/1.73 m2)	Δ Creatinine, μmol/L	Creatinine Baseline/Creatinine over 2 Days
1	66.16	79	124.1	37	57.94	1.88
2	52.70	88	103.4	45	50.70	1.96

) [16].

These two cases were identified by the treating physicians as adverse reactions to ketoprofen in the patients’ medical histories and were therefore analyzed by the investigators as a safety endpoint. Both cases were classified as category E according to the severity of harm of the National Coordinating Council for Medication Error Reporting and Prevention [17].

In addition to ketoprofen, patients received anticoagulants (heparin sodium, enoxaparin sodium), opioid analgesics (tramadol, morphine), antifibrinolytics (tranexamic acid), proton pump inhibitors (omeprazole), and antibiotics (cefazolin). Cefazolin was routinely administered to all included patients for surgical antibiotic prophylaxis.

Due to the small size of the AKI group and the absence of patients in certain subgroups, no comparisons were performed between patients with and without adverse reactions. Descriptive data are presented in

Table 7. Characteristics of Patients With and Without Acute Kidney Injury (AKI).

Indicator		AKI Patients (n = 2)	Patients Without AKI (n = 51)
Gender	Men (n, %)	0 (0)	22 (37.7)
	Women (n, %)	2 (5.7)	29 (52.8)
Alcohol Consumption	Yes	0 (0)	1 (1.9)
	No	2 (9.4)	50 (88.7)
Smoking	Yes	0 (0)	8 (15.1)
	No	2 (9.4)	43 (75.5)
Cardiovascular diseases, including coronary heart disease, hypertension, and congestive heart failure	Yes	2 (7.5)	35 (62.3)
	No	0 (1.9)	16 (28.3)
Type 2 Diabetes	Yes	0 (0)	7 (13.2)
	No	2 (9.4)	44 (77.4)
Chronic Kidney Disease	Yes	0 (1.9)	11 (18.9)
	No	2 (7.5)	40 (71.7)
Long-term use of NSAIDs prior to hospitalization	Yes	2 (9.4)	48 (84.9)
	No	0 (0)	3 (5.7)
ABCB1 rs1045642 Genotype	TT	2 (7.8)	16 (27.5)
	CC + CT	0 (2.0)	33 (62.7)
Age (years)		75.50 ± 2.12	65.80 ± 10.54
Body Mass Index (kg/m2)		31.10 ± 1.56	28.73 ± 5.45
Baseline Creatinine (μmol/L)		59.43 ± 9.52	86.12 ± 26.29
Baseline glomerular filtration rate CKD-EPI (mL/min/1.73 m2)		83.50 ± 6.36	73.82 ± 18.07
Duration of Surgery (min)		95.00 ± 14.14	74.71 ± 24.51
Total Protein on admission (g/L)		69.55 ± 4.88	66.28 ± 6.76
Creatinine over 2 days (μmol/L)		113.75 ± 14.63	85.52 ± 25.22
Glomerular filtration rate CKD-EPI over 2 days (mL/min/1.73 m2)		41.00 ± 5.66	72.45 ± 19.06
Total Protein at discharge (g/L)		59.55 ± 3.46	62.39 ± 4.76
Opioid Consumption During Hospitalization (MME)		20.00 ± 0	12.16 ± 5.41

MME—morphine milligram equivalents.

.

These two patients with acute kidney injury were followed after discharge. Their creatinine levels returned to the reference range within 7.8 ± 1.9 days post-discharge.

4. Discussion

We did not find significant associations between patient genotypes for the studied markers and pain levels or opioid analgesic consumption postoperatively when ketoprofen was used for pain management. Some associations were observed at 43 to 66 h after surgery for the rs10509681 CYP2C8 genotype groups—patients with TC and CC genotypes experienced more severe pain at rest and during movement compared with TT genotype carriers. However, this finding was made without correction for multiple comparisons and also was not replicated at other time points; therefore, its clinical significance is limited. We found no significant relationship between the efficacy or safety of postoperative pain management using ketoprofen and the CYP2C9 genotype. This finding is consistent with the classification of ketoprofen as a CYP2C9-independent NSAID, in accordance with CPIC guidelines. However, because the results are based on a small sample size and the observational nature of the study, they should be considered preliminary and warrant further investigation.

Furthermore, the study did not find any role of genetic polymorphisms encoding other enzymes, including UDP-glucuronosyltransferases, additional cytochrome P450 isoenzymes, and transporter proteins, in the response to ketoprofen for postoperative analgesia in patients undergoing total hip arthroplasty.

No chronic postoperative pain was observed after 6 months in our patients, and there were no requests for medical assistance during the year after surgery.

Two patients in our study developed acute kidney injury. Approximately 40% of in-hospital AKI cases are surgery-related [18]. Postoperative AKI is associated with risk factors such as older age, female sex, and severe comorbidities (chronic kidney disease, diabetes mellitus, hypertension, cardiovascular, liver, and pulmonary diseases), as well as obesity (BMI > 40 kg/m²), metabolic acidosis, type of surgery, and certain medications [19]. AKI patients in our study were older; however, comparison between groups was not valid due to the small number of cases.

Approximately 20% of hospital-acquired AKI cases are drug-induced. Nephrotoxic agents include NSAIDs, acetylsalicylic acid, various antimicrobials, contrast agents, diuretics, benzodiazepines, narcotics, psychotropic drugs, calcineurin inhibitors, cardiovascular medications, antiplatelet agents, statins, hypoglycemic agents, chemotherapeutics, and proton pump inhibitors [20]. Among the drugs mentioned above, all patients received ketoprofen and cefazolin. Additionally, some were receiving rosuvastatin and omeprazole therapy.

Pharmacogenetic studies on ketoprofen and dexketoprofen are limited. The relationship between the pharmacokinetics and safety of dexketoprofen and 46 polymorphisms across 14 genes (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, CYP3A5, UGT1A1, ABCB1, ABCC2, SLCO1B1, and SLC22A1) was investigated in 85 healthy volunteers from three bioequivalence clinical trials [8]. Some associations were identified between patient genotypes for CYP1A2, CYP2B6, and ABCB1 and the pharmacokinetic parameters of dexketoprofen; however, most were not confirmed in multivariate analysis.

No adverse reactions to dexketoprofen were reported in any of the three clinical studies, nor were there any clinically significant changes observed in laboratory or instrumental parameters. Consequently, no reliable association was found between pharmacogenetic markers and the pharmacokinetic parameters or safety of dexketoprofen. This is likely due to the minor role of cytochrome P450 enzymes and transporters in its metabolism, as well as the limited assessment of UDP-glucuronosyltransferase polymorphisms (only UGT1A1 was included).

In patients with rheumatoid arthritis and comorbidities, ketoprofen pharmacokinetics after a single 100 mg dose differed from that in healthy volunteers, particularly in terms of AUC [21]. The study included 11 patients who had comorbid conditions such as arterial hypertension, osteoporosis, type 2 diabetes mellitus, ischemic heart disease, chronic thyroiditis, and knee joint disease, for which they were receiving treatment. The authors noted that interindividual variability could not be explained by pharmacogenetics, as all genotyped patients were normal metabolizers for CYP2C8 and CYP2C9. The authors suggested that the observed decrease in AUC in patients with rheumatoid arthritis indicates the need for individualized ketoprofen therapy, taking into account other medications, patient age, and dietary habits.

In postoperative analgesia studies, associations between ketoprofen response and cytochrome P450 polymorphisms have been reported. In our previous study, CYP2C9*3 (rs1057910) was associated with higher analgesic efficacy of ketoprofen combined with tramadol on days 1–3 and day 5 after surgery, while rs1799853 of CYP2C9 was associated with more pronounced gastrointestinal symptoms [22]. These findings indicate associations between the efficacy and safety of postoperative analgesia with ketoprofen and patient CYP2C9 genotypes, suggesting that CYP2C9 gene polymorphisms may play a more significant role in the pharmacological response to ketoprofen than previously assumed.

In another study, CYP2D6 polymorphisms did not affect pain levels in patients receiving tramadol and ketoprofen, although sedation based on the Ramsay Sedation Scale differed between phenotypes [23]. However, this likely reflects the known role of CYP2D6 in tramadol metabolism rather than ketoprofen response.

Evaluation of ketoprofen pharmacokinetics together with pharmacogenetic and clinical data may improve study informativeness. Measuring drug concentrations at multiple time points could clarify variability in response and adverse events, although such designs are more complex and resource-intensive. Concomitant medications may further influence ketoprofen pharmacokinetics and complicate interpretation, highlighting the need to consider gene–drug interactions.

The limitations of this study include the following: most patients did not receive postoperative multimodal analgesia combining paracetamol and NSAIDs; instead, monotherapy with ketoprofen was used, with opioid analgesics prescribed only in cases of ineffectiveness.

Patients received additional analgesic therapy, including tramadol, morphine, or paracetamol, rather than ketoprofen, mostly during the first day after surgery. In this study, we hypothesized that ketoprofen largely influenced the level of pain and the need for opioid analgesics. However, the observed outcomes likely reflect the combined effect of the overall analgesic regimen rather than the isolated effect of ketoprofen alone, which limits the conclusions regarding ketoprofen pharmacogenetics. Also, the plasma concentration of ketoprofen was not measured in patients.

Additionally, the study was limited by a small sample size and the large number of polymorphisms analyzed, which may compromise the statistical power of the analysis.

The study is likely underpowered to detect modest pharmacogenetic effects. The number of patients in some genotype subgroups consisted of only a few individuals, making it difficult to interpret the results obtained. Larger studies and meta-analyses of several similar investigations are required to clarify the role of individual genetic polymorphisms in the response to NSAIDs, including ketoprofen.

Furthermore, the longitudinal pain data were analyzed using separate statistical comparisons rather than a repeated-measures approach due to missing diary entries. While we acknowledge that this increases the likelihood of chance findings (Type I errors), the overwhelmingly negative results suggest that this limitation does not affect our fundamental conclusion regarding the lack of genetic associations.

There were two cases of acute kidney injury during hospitalization which are insufficient to draw conclusions. The influence of pharmacogenetic factors on responses to other medications was not considered. Polymorphisms in genes such as CYP2D6, OPRM1, and COMT, which are associated with the response to opioid analgesics, were not included in the pharmacogenetic panel because these medications were prescribed on an as-needed basis, and their contribution was presumed to be insignificant.

5. Conclusions

We found no significant associations between the efficacy or safety of postoperative pain management and the genotypes of CYP2C9, nor with the genotypes of the other studied genes (CYP2C8, CYP3A4, CYP3A5, UGT2B7, ABCB1, PTGS1, PTGS2, C3orf20, and ZNF493-ZNF429) in patients receiving ketoprofen after total hip arthroplasty. Routine pharmacogenetic testing for ketoprofen is not supported by our findings.