Effects of Aspirin on Kidney Biopsy Bleeding Complications: A Systematic Review and Meta-Analysis (PROSPERO 2021 CRD42021261005)

Abstract

Postprocedural bleeding is the main complication of percutaneous kidney biopsy (PKB). Therefore, aspirin is routinely withheld in patients undergoing PKB to reduce the bleeding risk. The authors aimed to examine the association between aspirin use and bleeding during PKB. This systematic review and meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The article search was performed on MEDLINE and Scopus using queries specific to each database. Article inclusion was limited to primary studies. The meta-analysis compared the risk of major bleeding events between the aspirin-exposed versus nonexposed group. Pooled effect estimate was examined using random effects presented as odds ratio with 95% confidence intervals. Heterogeneity was assessed through Cochrane I² test statistics. Sensitivity and subgroup analyses were also performed according to kidney type. Ten studies were included in the review and four studies were included in the meta-analysis, reviewing a total of 34,067 PKBs. Definitions for significant aspirin exposure were inconsistent between studies, limiting comparisons. Studies with broader definitions for aspirin exposure mostly showed no correlation between aspirin use and postbiopsy bleeding. Studies with strict definitions for aspirin exposure found an increased risk of hemorrhagic events in the aspirin-exposed group. No significant differences were found between the aspirin-exposed and comparison groups regarding major bleeding events (odds ratio 1.72; 95% confidence interval 0.50 to 5.89, I²=84%). High-quality evidence on the effect of aspirin on the bleeding risk is limited. Our meta-analysis did not show a significantly increased risk of major bleeding complications in aspirin-exposed patients. Further studies are needed to define a more comprehensive approach for clinical practice.

Introduction

Before Iversen and Brun's initial description of kidney biopsy in the 1950s,¹ clinical and analytical findings were the basis for diagnosing kidney diseases.² Currently, kidney biopsies are performed in both native and graft kidneys, playing a crucial role in diagnosing many primary and secondary renal conditions.^3,4

Image guidance and automated biopsy devices have improved the diagnostic yield of kidney biopsies while simultaneously minimizing associated complications.^1,3–6 Despite advancements in the technique and equipment, major hemorrhagic complications are still responsible for lengthier hospital stays and increased treatment costs,^6,7 occurring in 0.24%–6.6% of all patients subjected to kidney biopsies.^2,4,6–23

Various factors have been associated with increased biopsy complication rates, including unmodifiable factors, such as age and sex, comorbidities (high serum creatinine levels, obesity, anemia, systemic autoimmune diseases, amyloidosis), and procedure-associated aspects.^{2,5–7,10,23–25} Most centers have similar safety requirements, with universal consensus regarding the need for adequate blood pressure control and correction of any preexisting coagulopathy.^26,27 Anticoagulants and antiplatelet drugs constitute other potential risk factors of hemorrhagic complications. Strategies regarding their preprocedural cessation are highly heterogeneous.

Aspirin has historically been considered a contraindication for percutaneous kidney biopsy (PKB). As one of the most widely prescribed antiaggregants worldwide,^28–30 its use is a common hurdle in patients proposed for PKB and determining whether and when to stop this drug before a biopsy assumes extreme relevance. Classical recommendations for a 7 to 10-day washout period were based on the premise that aspirin permanently inactivates cyclooxygenase, a key platelet enzyme for the synthesis of thromboxane.³¹ A daily dose of 30 mg aspirin is sufficient to completely suppress thromboxane production within 1 week.³² This can only be reversed by the generation of new platelets that have a mean life span of approximately 10 days.³³

Various entities have already debated this issue; however, a standardized and validated approach is still unavailable. The Society of Interventional Radiology recommends halting aspirin use at least 5 days before the biopsy.²⁷ The Canadian Cardiovascular Society recommends ceasing aspirin 7–10 days before the intervention.³⁴ The American College of Chest Physicians also suggests aspirin cessation 7–10 days before the intervention in low cardiovascular risk patients, but recommends maintaining aspirin use in moderate-to-high cardiovascular risk patients.³⁵ The Kidney Health Australia Caring for Australasians with Renal Impairment guidelines support aspirin use in high cardiovascular risk patients, but propose a late withdrawal strategy (3–7 days before the biopsy) in low-risk patients.³⁶

This article aims to review original articles regarding the influence of aspirin on the bleeding risk of patients subjected to PKB to help establish a standardized preprocedural strategy for its cessation.

Methods

This systematic review and meta-analysis was registered on PROSPERO (CRD42021261005) and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

Study Inclusion Criteria

Types of Studies and Study Participants

Randomized controlled trials and observational studies were considered eligible if they reported bleeding complications after kidney biopsy in patients under aspirin exposure. Case reports, narrative or systematic reviews, and meta-analyses were excluded. Studies dealing exclusively with dual antiplatelet therapy were excluded. All study participants were eligible, regardless of age, sex, underlying comorbidities, the presence of native or graft kidneys, or indications for kidney biopsy.

Types of Interventions

Studies were considered eligible for selection if kidney biopsies were performed percutaneously under image guidance: ultrasound (US) or computed tomography. Studies that included kidney biopsies performed using transjugular or other nonpercutaneous approaches were excluded.

Types of Outcome Measures

Bleeding complications comprised the primary outcomes. Major bleeding complications were defined as those resulting in permanent adverse sequelae or death or those requiring interventional therapy—medical (including blood transfusions), radiological, or surgical (such as embolization or coiling). Minor complications were defined as hematuria or hematomas requiring no active intervention or prolonged hospitalization and other nonsevere complications.

Data Source

Two different electronic databases were used, MEDLINE and Scopus. The following medical subject heading terms were used to identify potential articles: ”aspirin,” “acetylsalicylic acid,” and “antiplatelet” (Figure 1). Each medical subject heading term was then combined with “biopsy.” The authors agreed on a set of inclusion and exclusion criteria for automatic depuration. The search criteria were coded into queries specific to each database. The article search was conducted between July and August 2021. All studies published until the final article search were considered for inclusion.

Figure 1: Breakdown of obtained articles from the initial search included in the review, including queries used for each database.

Study Selection

Two investigators (JG and MR) independently evaluated the articles to determine their suitability for inclusion in this review. Eligible articles had their bibliographies examined to identify other relevant studies for this systematic review. Disagreements between the two reviewers were resolved by a third reviewer (LC).

Data Collection Process and Data Items

The following information—study type, type of kidney (native versus graft), number of kidney biopsies, type of image guidance, number of kidney biopsies performed in patients using aspirin, last known intake and dosage of the drug, clinical outcome, odds ratio (OR), and major limitations—was extracted from each selected article (when applicable) and condensed in Table 1. One author (JG) extracted the data, and the other author (MR) checked the information retrieved. Divergences over the interpretation of data were resolved by team discussion.

Table 1 - Study Characteristics.

Author (yr)	Study Type	Type of Kidney	No of Kidney Biopsies	Image Guiding Technique	No of Kidney Biopsies Performed in Patients Using Aspirin	Aspirin Cessation Date	Aspirin Dose	Clinical Outcome	OR	Major Limitations
Morgan et al., 2015 11	Observational retrospective cohort	Graft	235 a	US	59	NR	NR	No statistically significant association was found between aspirin use and the occurrence of bleeding complications.	OR 0.65	Retrospective design. Analysis of only a subgroup of patients for specific characteristics. No data regarding aspirin cessation date and doses used.
Baffour et al., 2017 13	Observational retrospective cohort	Graft	6700	US	1994	8–10 d: 664 biopsies 4–7 d: 855 biopsies 0–3 d: 475 biopsies	81 or 325 mg	325-mg dosage within 3 d of a biopsy was associated with an increased risk of bleeding. Other dosages and time interval exposure were not associated with an increase in the bleeding risk.	8–10 d: OR 0.405 4–7 d: OR 0.518 0–3 d: OR 2.52	Retrospective design. Arbitrary stratification of patient aspirin exposure into four groups may have confounded results. Possible higher level interaction effects between variables because of a low number of complications that limited the models used.
Kuiper et al., 2017 18	Observational prospective cohort	Graft	154	US	26	5 d: 25 biopsies Within 5 d: 1 biopsy	NR	Higher risk of a bleeding complication in patients who reported the use of aspirin.	OR 3.19	Only one cessation interval was evaluated. The patient who breached the aspirin cessation protocol was included in the same group for risk analysis. Small sample size.
Schwarz et al., 2005 20	Observational prospective cohort	Graft	1670	US	234	Sustained use	100 mg	Significant higher risk of gross hematuria. Nonsignificant increased risk for perirenal hematomas.	NR	No data regarding aspirin cessation dates for high-dose aspirin. Analysis for bleeding risk in aspirin-exposed patients was limited to those with sustained use at the time of the biopsy.
Mackinnon et al., 2008 15	Observational retrospective cohort	Native	1120	US	122	5 d: 54 biopsies Sustained use: 68 biopsies	NR	Higher risk of hemoglobin drop ≥1.0 g/dl (minor complication) in patients with concurrent antiplatelet therapy. No association was found with major complications.	NR	Retrospective design. Patients with exposures to different antiplatelet agents were pooled in the same group with no subanalysis regarding the bleeding risk associated with each specific drug.
Monahan et al., 2019 16	Observational retrospective cohort	Native	2204	US or CT	681	Within 10 d	81 or 325 mg	No significant associations between the variable and the outcome for any dosage.	325 mg: OR 0.778 Any dose: OR 0.612	Retrospective design. Exclusion of potentially clinically important complications below CTCAE 3. Low number of complications limited the models used. Potential confounding by periprocedural hypertension (not included in the regression model).
Lees et al., 2017 17	Observational retrospective cohort	Native	2563	US	327	Sustained use	75 mg	No increase in the risk of major bleeding nor the risk of hemoglobin drop >2 g/L or other morbidities in patients with concurrent aspirin use.	OR 0.6	Retrospective design. Small absolute number of complications. Underpowered, prospective confirmation would be required.
Atwell et al., 2010 14	Observational retrospective cohort	Native and graft	5832	US or CT	1270	Within 10 d	81 or 325 mg	No association was found between antiplatelet exposure and the risk of bleeding.	NR	Retrospective design. Exclusion of potentially clinically important complications below CTCAE 3. Use of a general database maintained over several years, with imprecise data regarding complications. Only one interval of aspirin exposure was evaluated.
Potretzke et al., 2019 19	Observational retrospective cohort	Native and graft	11,350	US or CT	NR	8–10 d or 4–7 d or 1–3 d or the day of the procedure	81 or 325 mg	Any aspirin usage within 3 d of the biopsy was associated with an increase in the risk of bleeding, particularly with a dosage of 325 mg. The association between taking aspirin on the day of the biopsy and the risk of bleeding complications was even more significant. Other time intervals for aspirin exposure were not associated with a higher risk of bleeding.	Renal biopsy (all) Any dose, day of biopsy: OR 12.4 325 mg, within 3 d: OR 9.57 81 mg, within 3 d: OR 2.1 Any dose, within 3 d: 3.66 Native kidney biopsy Any dose, day of biopsy: OR 5.75 325 mg, within 3 d: OR 5.57 Any dose, within 3 d: 5.51	Retrospective design. Potential confounders, such as underlying kidney or hepatic disease or use of other antithrombotic agents beyond clopidogrel and warfarin, were not considered. Aspirin usage data were dependent on patient self-reporting, which carries a risk of inaccuracy. Multiple biopsy operators with inherent variation in procedural techniques.
Bonani et al., 2021 21	Observational retrospective cohort	Native and graft	2239	US	37	Sustained use	NR	The risk to experience a major hemorrhage was higher in patients under aspirin.	OR 6.39	Retrospective design. Incomplete dataset. Patients under aspirin who stopped therapy a week before the biopsy were included in the same subgroup as those with no aspirin exposure.

NR, not reported; OR, odds ratio; US, ultrasound; CT, computed tomography; CTCAE, Common Terminology Criteria for Adverse Events.

^aAlthough a total of 2514 biopsies were considered in this study, only 235 biopsies were included in the subgroup analysis that allowed the authors to study the relation between aspirin exposure and the development of bleeding complications.

Assessment of the Risk of Bias in the Included Studies

The risk of bias was independently evaluated by two reviewers (JG, MR) using the Risk Of Bias In Non-randomized Studies of Interventions tool. Seven different bias domains were evaluated independently for each selected article—confounding bias, selection bias, classification bias, performance bias, missing data, measurement bias, and reporting bias. Disagreements were resolved by a third reviewer (LC) or consensus-based discussion.

Data Analysis

We conducted a meta-analysis to quantitatively assess the risk of major bleeding complications between aspirin-exposed versus nonexposed groups. In the exposed group, we included PKB performed in patients with aspirin exposure within 3 days of biopsy. The nonexposed group included biopsies performed in patients with no aspirin exposure at least 7 days before this procedure. Meta-analyses were performed using Review Manager version 5.3 (Cochrane Collaboration). Dichotomous outcome results were expressed using OR with a 95% confidence interval (CI). A P-value of <0.05 was considered significant. Data were pooled using a random-effects model (Mantel-Haenszel method) to incorporate both within-study and between-study variances. Statistical heterogeneity among studies was assessed using I² tests with substantial heterogeneity defined as I² >50%.³⁷ Sensitivity analyses were performed to explore potential sources of heterogeneity, including removal of studies with a non–low risk of bias for each domain assessed and the consecutive removal of each individual study. We also conducted a subgroup analysis according to kidney type (native or graft). Publication bias was assessed through visual inspection of the funnel plot.

Results

Study Selection

Figure 1 presents the study selection flow of this systematic review. In total, 8204 articles were identified through electronic searches in the two selected databases. Automatic depuration eliminated 4129 articles. Of the remaining 4075 articles, 4064 were found ineligible for inclusion after having their titles and abstracts screened. Two of the remaining articles were duplicates. After scanning the remaining ten articles' bibliographic references, three additional studies were found to fall in the scope of the review, meeting the requisited criteria. The methodology of each article was then assessed to determine suitability for our analysis. Three articles were found to have inadequate designs for the proposed analysis (Reschen et al.,³⁸ Ori et al.³⁹ and Trajceska et al.¹²). The remaining ten^11,13–21 articles were included in our analysis. Four^11,13,17,21 of the ten studies selected were also included in quantitative synthesis.

Study Characteristics

Details of the included trials, such as study design, type of kidney biopsied (native versus graft), total number of biopsies included, number of biopsies performed under aspirin exposure, image technique used, criteria for aspirin exposure, aspirin doses used, and clinical outcomes related to bleeding complications, are provided in Table 1.

This systematic review included studies conducted between 2002 and 2019, with an aggregate 34,067 PKBs. Sample sizes ranged from 154 to 11,350 PKBs. Four studies related exclusively to kidney graft biopsies,^11,13,18,20 three only included native kidney biopsies,^15–17 and three others discussed complications in both native kidneys and grafts.^14,19,21 All biopsies were performed under image guidance—in all but three studies (which also included computed tomography–guided biopsies),^14,16,19 US was the only imaging modality used. All studies were observational cohort studies, and only two were prospective in nature.^18,20

The definition used for aspirin exposure was disparate. While some studies limited valid aspirin exposure to a confirmed dose the same day as the kidney biopsy, others had broader criteria that included patients with any confirmed dose within 10 days of the intervention.^14,16 In other studies, different time intervals were considered,^13,18,19 whereas some did not objectively define the date of the last confirmed dose.¹¹ The definition of bleeding complications was more consistent among the selected studies.

Risk of Bias within and across Studies

The authors used the ROBINS-I tool to estimate the risk of bias of each included article across seven domains. The risk of bias assessments is presented in Figures 2 and 3. Most of the studies included in this review were not specifically designed to evaluate the correlation between aspirin use and bleeding complications. As such, in most of the included articles, solid data regarding the time frame of aspirin withdrawal is scarce and appropriate analysis for confounding factors is absent. Thus, all articles were found to be at moderate-to-serious risk of overall bias.

Figure 2: Risk of bias summary (Risk Of Bias In Nonrandomized Studies of Interventions tool).

Figure 3: Risk of bias graph (Risk Of Bias In Nonrandomized Studies of Interventions tool).

Synthesis of Results

Kidney Graft Biopsies

A total of 11,038 biopsies were assessed in the articles related to kidney grafts, including elective (or surveillance) and indication biopsies.

Morgan et al.¹¹ analyzed 2514 graft biopsies. A major complication of a graft biopsy was defined as the need for additional hospitalization time and/or an active intervention, including hospital admission, transfusion of blood products, interventional radiology vascular procedure, surgical exploration, transplant removal, and/or death. Minor complications were not evaluated in this study. Major complications occurred in 47 patients (1.9%). Although patients were advised to stop both anticoagulants and antiplatelet agents 7 days before the biopsy, that was not always feasible (particularly in the case of indication biopsies). To study the relationship between aspirin and the risk of biopsy complications, the authors selected all patients with a biopsy complication and a subset of patients without any complication using a 1:4 ratio, forming a subgroup of 235 individuals (47 patients with major complications and 188 patients with no complications). Nine patients met the article's aspirin exposure criteria in the complication-afflicted fraction (19.15%) while 50 patients had relevant aspirin intake in the uncomplicated fraction (26.60%). The percentages stated in the original article (15% versus 85% aspirin exposure in the complicated and uncomplicated subgroups respectively) were erroneously calculated using the total number of patients in the aspirin-exposed subgroup as the whole, meaning the percentages solely represent the 1:4 ratio selected for the construction of the control sample. Univariate analysis suggested aspirin use was not significantly associated with biopsy complications (including hemorrhagic phenomena) (see Table 1). In addition, the authors compared the risk of complications between indication biopsies (n=1599) and protocol biopsies (n=915). The first were associated with a significantly higher complication rate than the latter (2.7% versus 0.33%, P-value<0.001).

Baffour et al.¹³ analyzed the association between aspirin use, its dosage (81 or 325 mg), and the development of hemorrhagic complications in 6700 kidney graft biopsies. Most of the biopsies were performed in patients without aspirin exposure or with a last registered dose more than 10 days earlier (n=4706). Six hundred sixty-four biopsies were performed within 8 to 10 days of the last dose; 855 others within 4–7 days; and on 475 occasions, the patients had taken aspirin within 3 days of the procedure (n=475). The authors analyzed two outcomes: general and major bleeding complications, with events graded on the basis of the SIR Standards of Practice complication classification system (minor complications: A—requiring no therapy or consequence; B—requiring nominal therapy with no consequence; major complications: C—requiring therapy or minor hospitalization; D—requiring major therapy, unplanned increase in level of care, or prolonged hospitalization; E—resulting in permanent adverse sequelae; F—resulting in death). For both outcomes, adjusting for multiple testing, only a dosage of 325 mg within 3 days showed a significant association (OR=3.87, P-value=0.032 for any bleeding complication; OR=6.30, P-value=0.024 for major bleeding complication). However, aspirin in any dosage and up to 7 days before the biopsy showed a clear trend toward more major complications (e.g., OR of 4.39, P=0.010 for aspirin in any dosage within 7 days of the biopsy). The authors also compared the risk of general and major bleeding complications between protocol biopsies (80.3%) and indication biopsies (19.7%). For both events, indication biopsies presented a statistically significantly higher risk (OR=2.27, P-value=0.012 and OR=6.84, P-value<0.001, respectively).

The prospective study by Kuiper et al.¹⁸ evaluated 154 transplanted patients who underwent graft biopsy. The authors aimed to determine whether bleeding tests could predict bleeding after kidney biopsy. Bleeding outcomes included hemoglobin drops of 1 mmol/L or more, positive US findings of a perinephric and/or subcapsular hematoma, and de novo hematuria (>200 red blood cells per high power field) lasting over 24 hours. Participants needing postbiopsy intervention (e.g., embolization, coiling, explantation, transfusion of packed cells, bladder catheterization or other) were marked having a major bleeding complication. Of the selected patients, 26 were chronically medicated with aspirin—all but one ceased aspirin use 5 days before biopsy. Patients who reported aspirin use had a significantly higher risk of bleeding complications in comparison with those without aspirin exposure (OR=3.19) (see Table 1).

Schwarz et al.²⁰ performed 1171 protocol biopsies and 499 indication biopsies. The authors aimed to evaluate the safety and adequacy of protocol biopsies after kidney transplantation. Bleeding outcomes included gross hematuria and perirenal hematoma. Patients under sustained low-dose (100 mg) aspirin exposure (n=234) had slightly more often gross hematuria and perirenal hematoma than patients without (gross hematuria, 5.8% versus 3%, P=0.046; perirenal hematoma, 3.2% versus 2.2%, not significant). Other aspirin cessation dates were not considered for analysis.

Native Kidney Biopsies

The selected studies on native kidney biopsies included a total of 5887 procedures. The association between bleeding events and aspirin use was inconsistent.

Monahan et al.¹⁶ analyzed 2204 native kidney core biopsies. Thirty-one percent of the patients had taken aspirin within 10 days of the biopsy. Only major adverse bleeding events were included in the analysis (Common Terminology Criteria for Adverse Events [CTCAE] grade 3 or higher). Specifically, a grade 3 event includes the need for blood transfusion or radiologic, endoscopic, or elective operative intervention. A grade 4 event has life-threatening consequences and needs urgent intervention. A grade 5 event is death. The authors found no significant association between aspirin exposure and bleeding complications (OR=0.778, P-value=0.732 for 325 mg; OR=0.612, P-value=0.222 for any dosage).

Lees et al.¹⁷ analyzed 2563 biopsies, including 1499 elective and 1064 emergency procedures. Three hundred twenty-seven patients reported aspirin use (75 mg) and were allowed to continue the drug despite the biopsy. Both minor (patients with hemoglobin drops >2 g/L, visible hematuria, requirement for catheter/irrigation, documented hematoma, additional period of observation or loin pain) and major (patients requiring blood transfusion, surgical or radiological intervention or death) bleeding complications were reported. However, analysis was limited to major bleeding events, and the authors admitted that minor bleeding events may have been underreported. No association was found between aspirin use and major bleeding complications (2.5% versus 1.6%, P-value=0.47). In patients taking aspirin, there was no significant increase in the risk of a hemoglobin decrease of >2 g/dl (2.4% versus 2.2%, P-value=0.79) or other morbidities (1.5% versus 3.0%, P-value=0.15). However, the authors found a significantly increased risk of major bleeding in emergency biopsies compared with elective procedures (3.4% versus 1.1%, P-value<0.001).

Mackinnon et al.¹⁵ included two centers with different policies regarding aspirin use in patients undergoing kidney biopsy. One thousand one hundred twenty biopsies were retrospectively analyzed. In one center (n=523), all patients using antiplatelet agents (n=60) were asked to cease their use 5 days before the biopsy. In the other center (n=597), patients using antiplatelet agents (n=75) maintained the medication. Both minor (defined as a fall in hemoglobin of ≥1.0 g/dl, without the need for transfusion or intervention) and major (patients requiring postbiopsy intervention—including transfusions, embolization, or placement of an irrigation catheter) bleeding complications were analyzed. The risk of a hemoglobin drop greater than 1.0 g/dl was significantly higher in the center where patients were kept under antiplatelet agents (31% versus 11.7%; P-value=0.008). However, no association was found between concurrent antiplatelet agent use at the time of the biopsy and major complications (1.3% versus 0%; P-value=0.56) (see Table 1).

Native and Graft Kidney Biopsies

The selected studies regarding both native and graft kidney biopsies included 19,421 biopsies.

Atwell et al.¹⁴ retrospectively analyzed 15,181 percutaneous biopsies, including 5832 kidney biopsies—1270 of which were performed under aspirin exposure, defined as the use of aspirin (81 or 325 mg) within 10 days of the biopsy. Only major bleeding events were included in the analysis (CTCAE grade 3 or higher). The authors found no significant difference on the bleeding risk of patients with documented aspirin exposure (1.0% versus 0.6%, P-value=0.53) (Table 1).

Potretzke et al.¹⁹ retrospectively analyzed 3245 native kidney and 8105 graft biopsies. Aspirin exposure was categorized according to the dose and cessation date. Last intake dates were divided into the following categories: 8–10 days before the biopsy, 4–7 days before the biopsy, 1–3 days before the biopsy, and the day of the procedure. The authors found that, for any studied dose, usage within 3 days of the biopsy was associated with an increased bleeding risk (OR=3.66, P-value <0.001)—this was particularly evident in the group treated with the highest aspirin dosage (325 mg). Only major bleeding events were included in the analysis (CTCAE grade 3 or higher). Aspirin use on the day of the biopsy was shown to have a particular effect on the incidence of postbiopsy bleeding complications (OR=12.4, P-value<0.001). Other time intervals for aspirin exposure were not associated with a higher risk of bleeding after the biopsy.

Bonani et al.²¹ retrospectively analyzed 733 native kidney and 1506 graft biopsies. The authors aimed to evaluate the safety of kidney biopsies when performed as an outpatient procedure. All bleeding events were reported, but the influence of aspirin use was only evaluated for major bleeding events. Thirty-nine patients were submitted to kidney biopsy under concurrent aspirin use. An undefined number of patients discontinued antiplatelet therapy a week before the biopsy and were considered unexposed to aspirin. Univariate analysis showed an increased risk of major hemorrhage in patients under aspirin (10.3 versus 1.8%; P<0.001). Multifactorial logistic regression was performed to identify independent predictors of bleeding. Two models were used: one incorporating only clinical parameters that were available for most patients (model 1) and one incorporating laboratory parameters and blood pressure, which were available for only a subset of patients (model 2). Aspirin use was significantly associated with major bleeding in model 1, but lost significance in model 2.

Quantitative Synthesis

Major Bleeding Events

Owing to significant heterogeneity in methodology, definitions of aspirin exposure, and reported outcome measures, only four^11,13,17,21 of the ten included studies presented data considered compatible with statistical combination.

The primary analysis is described in Figure 4. The aspirin-exposed group included 901 kidney biopsies, and the nonexposed group included 8924 biopsies. The summary OR of major bleeding events was 1.72 (95% CI 0.50 to 5.89, P=0.39) with significant heterogeneity among studies (I²=84%; P=0.0003).

Figure 4: Forest plot for the major bleeding events between aspirin-exposed and nonexposed groups. CI, confidence interval.

Sensitivity analyses were performed to explore potential sources of heterogeneity, including removal of studies with a non–low risk of bias for each domain assessed and the consecutive removal of each individual study (see Supplemental Tables 1 and 2). No significant difference in the overall effect estimates was found. Further subgroup analysis on the basis of the kidney type (native or graft) also showed no significant effect on the pooled estimate and 95% CI (Figure 5). The funnel plot did not show any obvious asymmetry (Supplemental Figure 1).

Figure 5: Subgroup analysis for major bleeding events between aspirin-exposed and nonexposed groups based on kidney type (native or graft). CI, confidence interval.

Discussion

This study was designed to ascertain the risk of bleeding complications associated with aspirin exposure in patients subjected to PKB. Ten studies fulfilled eligibility criteria, and an aggregate 34,067 kidney biopsies were included. Our meta-analysis showed no significant differences regarding major bleeding events between the aspirin-exposed and nonexposed groups (OR 1.72; 95% CI 0.50 to 5.89, I²=84%).

Kumar et al.²⁵ also reviewed data on the risk of PKB complications under aspirin use. Despite a wider set of inclusion criteria, with guidelines and systematic reviews included in their analysis, the authors lamented the lack of quality evidence and supported the guideline recommending a 7 to 10-day washout period in most elective procedures. However, this review did not include the more recent studies by Baffour et al.¹³ and Potretzke et al.¹⁹ with timeframe-dependent subgroup analysis. The inclusion of these two articles is a major strength of our study.

From the selected articles, considering their design and limitations, Baffour et al.¹³ and Potretzke et al.¹⁹ were found to present more robust data, particularly because of the inclusion of a timeframe-dependent subgroup analysis. In these studies, aspirin use on the day of the biopsy, particularly in higher dosages, was found to facilitate postbiopsy bleeding complications. However, aspirin cessation 3 days before PKB was found equivalent to more conservative strategies, with no increased risk of bleeding events. Lees et al.¹⁷ and Mackinnon et al.¹⁵ reported no significant association between aspirin use and major bleeding complications in native kidney biopsies, despite an apparent trend in the latter study toward more complications under antiplatelet agent exposure that could arguably become significant with a greater sample size. This trend was also apparent in our meta-analysis, but owing to the high heterogeneity among the enrolled studies and the uncertainty associated with the wider CI, the presented results should be interpreted with great caution.

Evidence supporting deferred aspirin withdrawal is also relevant because of the increased risk of cardiovascular complications found after aspirin interruption,^40–47 partly explained by rebound phenomena (e.g., increased thromboxane synthesis).^41,48 This is particularly relevant in patients with established cardiovascular disease where aspirin is used for secondary prevention. The risk of myocardial infarction may increase up to 60% in this population,⁴⁹ and acute coronary syndrome may occur in less than two weeks from the withdrawal of aspirin.⁵⁰ This average delay between aspirin withdrawal and myocardial infarction is consistent with the rebound in platelet activity after aspirin cessation.³³ Surprisingly, the outcomes of patients who withdrew aspirin are worse than users or previously nonusers, with an increased risk of death.⁵⁰ Shorter aspirin cessation strategies may reduce cardiovascular events because of a more limited aspirin-free window, additionally benefitting from a quicker indication-to-intervention interval, meaning the diagnostic intervention is performed sooner, thus anticipating the introduction of subsequent therapeutic measures. In fact, there is a 30% recovery of platelet function 48 hours after aspirin administration,³¹ which could be enough to ensure a clinically relevant bleeding risk reduction.

The lack of consistent data on the complications of kidney biopsy associated with aspirin leads us to speculate that its periprocedural management should be performed individually, considering clinical and technical aspects. The threshold for aspirin withdrawal is likely to be different in patients who combine good anatomical features and high cardiovascular risk compared with those with an anatomically demanding kidney and low cardiovascular risk. In addition, other aspects, such as the quality of the US image, can also modify the risk of complications and consequently influence the decision to discontinue aspirin. All these individual conditions are difficult to incorporate into guideline recommendations.

Our study presents several limitations. First, most of the included studies had an observational and retrospective design. Second, significant heterogeneity in methodology, participant selection, and reported outcome measures limited the quantitative comparisons. Heterogeneity was particularly evident in the criteria used for aspirin exposure, varying from a confirmed dose the same day as the kidney biopsy^15,17 to any confirmed intake within 10 days of the intervention,^14,16 with some authors failing to define relevant aspirin exposure.¹¹ Moreover, some studies^{11,13,14,16,19} focused primarily on major adverse bleeding events, meaning minor bleeding complications related to aspirin use may have been underreported.

Additional well-designed prospective controlled trials are needed for a better comprehensive understanding of this association. For future research, we suggest authors should focus not only on the bleeding complications related to aspirin exposure but also on the clinical outcomes associated with aspirin discontinuation, namely thrombotic complications and cardiovascular events. This would allow clinicians to make better informed decisions with a more substantiated understanding of the risks and benefits of aspirin cessation.

In patients awaiting PKB, the best strategy regarding aspirin cessation is still undetermined because of lack of high-quality evidence. Our meta-analysis did not show a significantly increased risk of major bleeding complications in aspirin-exposed patients.