Abstract
Background
While the majority of bleeding complications after a percutaneous kidney biopsy (PKB) occur early (≤24 h), delayed onset bleeding complications (>24 h) have been rarely reported and can be catastrophic for the patient.
Purpose
To describe the incidence, risk factors, and outcomes of delayed bleeding complications after PKB.
Material and Methods
We retrospectively studied native and graft kidney biopsies in patients who developed delayed bleeding complications (>24 h) after the biopsy performed in the Department of Nephrology and Renal Transplantation of a tertiary care medical institution in north India between January 2014 to December 2018.
Results
Of the 4912 renal biopsies reviewed, 20 patients (16 men, 4 women; 0.40%) had a delayed biopsy bleeding complication. Of these patients, 95% had major bleeding complications requiring blood transfusions and 85% needed intervention like gelfoam/coil embolization. Despite intervention, one patient (5%) had mortality due to complications of bleeding and sepsis. When compared to a control group of patients with early biopsy bleed, patients with the delayed biopsy bleed had similar demographic and clinical profiles except for higher pre-biopsy hemoglobin and lower systolic and diastolic blood pressure.
Conclusion
A post-PKB delayed onset bleed is not uncommon, and the vast majority of these patients had major bleeding complications requiring blood transfusions and/or intervention like embolization. They had a similar demographic and clinical profile presentation as early bleed patients. Meticulous outpatient monitoring and patient education after discharge may be useful to detect this complication promptly and to intervene early to have good patient outcome.
Keywords
Introduction
A percutaneous kidney biopsy (PKB) is one of the most vital and an indispensable tool in the armamentarium of nephrologists for the diagnosis and prognosis of various renal primary and secondary diseases (1). The PKB technique has evolved in the pursuit of safety from the use of the modified Vim-Silverman needles and sitting position in the 1950s by Karl and Muehrcke to the prone position of patients, real-time ultrasound guidance, and use of automated spring-loaded biopsy guns in the current era (2). However, PKB is not possible without its complications. The most common complication is bleeding, which can be catastrophic and organ- or life-threatening to some patients. Hence, identifying the rate of bleeding complications, its timing, modifiable and non-modifiable risk factors, and ideal post-biopsy assessment protocol is essential information for nephrologists and patients before providing informed consent.
PKBs are currently performed by both nephrologists and radiologists and many perform it as an outpatient procedure in selected patients. However, there is a debate on the appropriate length of in-hospital observation after the procedure and the safety of outpatient PKB is a subject of contention. A large study by Whittier et al. (3) found that 89% of patients with bleeding complications were identified in ≤24 h. A recent study by Atwell et al. (4) showed an incidence of 0.28% of delayed bleeding occurring after 24 h.
While there are few case reports of delayed post-PKB bleeding, there is no published study so far that has examined the demographic, laboratory, or clinical risk factors other than the conventional risk factor that predicts delayed bleeding after PKB and its outcomes. The aim of the present study was to examine the incidence of delayed post-PKB bleeding complications, predictive risk factors other than conventional risk factor, and its outcomes.
Material and Methods
This was a retrospective single-center observational study conducted at tertiary care medical institute in north India between January 2014 and December 2018.
Inclusion criteria
Patients who had a complication after PKB were identified from the electronic medical records (EMR), medical note sheet, and kidney biopsy registers.
Delayed biopsy bleed was defined in our study as patients who had a bleeding complication occurring >24 h after PKB.
The age- and sex-matched “control group” was selected (in a 1:2 case:control ratio) among patients who underwent PKB at the same center within the same period and had bleeding complications identified in the first 24 h of the biopsy.
Exclusion criteria
Patients who did not develop any bleeding complications were excluded from the present study.
Patients’ demographic data, indication for biopsy, all laboratory reports, and clinical and hemodynamic parameters were collected from the EMR and their individual files. Readmissions, if any, and mortality data were also collected from the EMR.
Protocol of procedure of renal biopsy
Informed consent was obtained from all patients included for renal biopsy and had normal coagulation parameters (bleeding time was not routinely measured at our center) and platelet count >100 × 103/μL. Anticoagulation and antiplatelet drugs are routinely withdrawn seven days before PKB at our center. Blood pressure should be less than 140/90 at the time of biopsy.
The PKBs were performed with real-time ultrasound imaging and needle tracking guide (Ultra-pro II™ Needle guide; Civco, IA, USA) on a Toshiba, Japan, Xario model SSA-660a, with a 3.75-MHz curvilinear probe. All biopsies were performed by a nephrologist and a trainee nephrologist. The biopsies were taken using automated spring-loaded devices (BARDR Max-coreR disposable core biopsy needle; Bard Peripheral Vascular Inc., USA) with a 16/18 gauge and 16-cm long needle with a penetration depth of 22 mm and sample notch of 18 mm.
All patients were advised to lay supine in the bed with a compression bandage over the biopsy site for at least 6 h with overnight bed rest and close monitoring of urine for gross hematuria, blood pressure, and other vitals. Hemoglobin and hematocrit were measured 6 h and 24 h after the biopsy. A screening ultrasound was done by the nephrologist performing the biopsy immediately after the biopsy. Any patient with clinical features to suggest a bleeding complication like gross hematuria, persistent flank pain, or hemodynamic instability and significant hemoglobin decline after the kidney biopsy underwent an ultrasound or computed tomography (CT) evaluation of the abdomen by radiologists.
Outcome measures
The incidence, time to detection, and histological diagnosis of delayed biopsy bleeding complications among all biopsies were observed. Demographic, clinical, and coagulation parameters were compared between the delayed biopsy bleed group and an early bleed group to find out any risk factors. The incidence of minor and major bleeding complications in early and delayed bleeding groups and its management was observed between the two groups.
Minor complications were defined as perinephric hematoma and gross hematuria that resolved spontaneously without intervention and did not require a blood transfusion.
Major complications included a significant decline in hemoglobin level requiring blood transfusions, arteriovenous fistula or renal arterial pseudoaneurysm requiring radiological intervention like angiography, gelfoam transarterial embolization or coil embolization, nephrectomy, and death due to the bleeding complication.
Statistical analysis
Unpaired Student’s t-test was used for comparative analysis between two groups. Pearson’s chi-square test/Fisher’s exact test was employed to analyze the categorical data as appropriate. P < 0.05 was considered significant. Statistical package for social sciences, version 22 (IBM Corp., Chicago, IL, USA) was used for statistical analyses.
Results
Baseline characteristics
From January 2010 to December 2018, 4912 kidney biopsies were performed at our center. Delayed post-renal biopsy bleed was seen in 20 patients (16 men, 4 women; 0.40%). Out of the 20 patients with delayed bleeding complications, 19 had undergone native kidney biopsies and one patient had a graft biopsy. Baseline characteristics of the delayed post-renal biopsy bleed groups were compared with a select short cohort from the early bleed group in Table 1.
Comparison of baseline characteristics of delayed and early post biopsy bleed groups.
Values are given as mean ± SD unless otherwise specified.
aPTT, activated partial thromboplastin time; BMI, body mass index; INR, international normalized ratio; PT, prothrombin time.
The demographic and baseline clinical parameters of the early and delayed bleeding complication groups were compared and no statistically significant difference was found in their age, gender, platelet count, international normalized ratio, activated partial thromboplastin time, serum creatinine, pre-biopsy desmopressin use, and plasmapheresis after the kidney biopsy. However, patients with the delayed biopsy bleed had significantly higher pre-biopsy levels of hemoglobin and lower systolic, diastolic, and mean arterial blood pressure at presentation as compared to patients who presented with early biopsy bleed.
Clinical presentation
The average time of detection of delayed biopsy bleed was 7.6 ± 4.1 days. The management and outcomes of early and delayed bleed are compared in Table 2.
Comparison of management and outcomes of delayed and early post biopsy bleed.
Values are given as n (%) or mean ± SD.
The patients with delayed bleed had a statistically higher incidence of major complications requiring blood transfusions and intervention procedure including gelfoam, coil embolization or both (17/20 vs. 7/43, P < 0.001 for both). In the delayed biopsy group, the intervention procedure, including coil embolization, gelfoam, and both coil and gelfoam embolization, was done in seven, six, and four patients, respectively. In the delayed biopsy bleed group, 5 (25%) patients were hemodynamically unstable and 4 (20%) needed readmission with bleeding after discharge compared to none in the early biopsy bleed control group (P = 0.001 and 0.0034, respectively). None of the patients in either group underwent nephrectomy to control the bleeding.
These patients in the delayed bleed group also had a longer total duration of stay in the hospital (23.6 ± 13.7 vs. 15.7 ± 7.5, P = 0.005).
One patient in the delayed biopsy group died after readmission. Although the patient’s bleeding had stabilized after the intervention, there were complications with sepsis and multi-organ dysfunction as the patient was on oral steroids for the management of IgA nephropathy.
Etiological diagnosis and management of patients with delayed bleed
Table 3 describes the individual patients with delayed biopsy bleed complications, the time to detection of the bleed, and outcomes. All the patients with delayed biopsy bleed had glomerular diseases on the kidney biopsy, the most common being Ig A nephropathy (7/20, 35%) followed by unspecified diffuse global glomerulosclerosis (5/20, 25%), amyloidosis (2/20, 10%), Paucimmune crescentic glomerulonephritis (2/20, 10%), and one case each of antibody-mediated rejection, membranoproliferative glomerulonephritis, diffuse proliferative glomerulonephritis, and lupus nephritis class IV.
Delayed biopsy bleed complications (age, gender, time to bleed, intervention, histological diagnosis and final outcomes).
*Radiological intervention to control bleeding.
ABMR, acute antibody mediated rejection; AVF, arteriovenous fistula; CrGN, crescentic glomerulonephritis; DGGS, diffuse global glomerulosclerosis; DPGN, diffuse proliferative glomerulonephritis; DSA, digital subtraction angiography; GN, glomerulonephritis; LRTI, lower respiratory tract infection; MPGN, membrano-proliferative glomerulonephritis.
Discussion
PKB is a well-established investigation for the accurate diagnosis of kidney diseases. Though usually safe, kidney biopsy can be associated with serious complications. The risk of minor bleeding episodes after PKB including self-limiting gross hematuria or small hematomas that did not need any blood transfusion or intervention is in the range of 17%–33% in a large meta-analysis of 9474 kidney biopsies, whereas major complications, including hemorrhage requiring transfusion, angiographic intervention, nephrectomy, or death, have been reported in only around 2%–8% of renal biopsies (5,6) Predictors of major bleeding are the presence of coagulopathy, kidney dysfunction (serum creatinine, >2.0 mg/dL), low concentration of hemoglobin, use of a wide needle, or atherosclerosis (7,8).
The current data are from the only available study in the literature, on the issue of delayed onset bleed after PKB. In the present study, we have compared the clinical and demographic profiles of delayed bleeding with early bleeding and observed no difference between two groups. The predictors of major bleeding between the two groups were also the same, such as coagulation disturbances, renal function >2.0 mg/dL, low level of hemoglobin, and use of wide bore needles. However, these patients with delayed bleed had a statistically higher incidence of major consequences requiring more blood transfusions, more intervention procedures, more hemodynamic instability, more readmissions, and longer hospital stay in comparison with patients with early bleed.
In the present study, the incidence of delayed post biopsy bleed was 0.40%. This incidence is twice higher than that reported from a large retrospective study by Atwell et al. (4) of solid organ image-guided percutaneous biopsy where the delayed (>24 h) bleeding complication rate of native renal biopsy was 0.28%. This difference in the incidence of delayed biopsy could be because of the larger sample size of kidney biopsy in the present study (4912 vs. 1407).
In the present study, none of the conventional risk factors for post-kidney biopsy bleeding were found to be associated with an increased risk of delayed biopsy bleed hence making it difficult to predict clinically. One of the earliest case reports of delayed post-biopsy bleed was reported in 1959, where the patient developed flank pain, hypotension, and hematoma nine days after the kidney biopsy and required nephrectomy to control the bleeding (9). Another case of delayed biopsy bleed was reported in a patient with diabetic nephropathy class IIa that bled after seven days and that was treated with angiographic coil embolization (10). Traditional predictors of major bleeding such as coagulopathy, kidney dysfunction (serum creatinine <2.0 mg/dL), low concentration of hemoglobin, or the use of a wide needle (i.e. 14 gauge), were not identified in this patient.
The time between the kidney biopsy and detection of the delayed bleeding complication in the present study was in the range of 3–21 days and the mean time period was 7.6 ± 4.1 days. As it is impractical and expensive to delay discharge after a kidney biopsy for such extended periods, a recommendation for increasing the period of inpatient observation after a biopsy cannot be made. Further, none of the usual risk factors that predict post-biopsy complications were found in our study to be predictive specifically for higher risk of delayed bleeding complications. None of the patients reported trauma or strenuous physical activity. Hence, further studies are needed to elaborate on these risk factors and predictive modelling.
The vast majority of delayed bleed patients in the present study needed a blood transfusion and intervention in the form of angiography with gelfoam and coil embolization, yet they were safely managed like the earlier published case reports (11,12) and study by Jiang et al. (13). The study by Jiang et al. found that four of six major complications occurred within 4 h of the biopsy and delayed major complications which occur at 12 h and 62 h after biopsy were safely managed with interventions.
Of the 20 patients, 2 (5%) in the delayed bleeding group had amyloidosis, which is higher than the incidence of amyloidosis on kidney biopsy at our center of 1.2%. This is consistent with the higher frequency of bleeding complications in patients with amyloidosis found in a study by Eiro et al. (14). This increased risk is due to amyloid angiopathy with increased fragility of blood vessels and acquired hemostatic deficiencies (15).
There was no significant difference in plasmapheresis rates between the early and late biopsy bleed groups in the present study. At our center, plasmapheresis is usually withheld until 36–48 h after the renal biopsy unless a pressing indication is present and potential benefits outweigh risks. A case report published by Hickstein et al. (16) described a patient with pauci immune glomerulonephritis with microscopic polyangiitis treated with plasmapheresis that developed a delayed biopsy bleed after 11 days of the biopsy.
Table 4 summarizes various case reports and case series in the literature and our data on the issue of delayed onset bleed after PKB. A total of 30 patients were included in the present data and the incidence of delayed bleed is in the range of 0.28%–0.4%. The timing of delayed post-PKB is in the range of 3–21 days. One out of 30 delayed bleed patients underwent nephrectomy. Two out of 30 (6.66%) delayed bleed patients died.
Comparison of different case reports/literature on delayed biopsy complication (>24 h).
*Case reports.
NA, not available.
The strength of the present study is that this is the only study available on the issue of delayed onset post-PKB bleed complication in the literature especially in light of upsurging evidence of early discharge after PKB.
The limitations of the study are its retrospective nature and small number of cases. Hence, we propose prospective studies to evaluate predictive risk factors, ideal monitoring, and follow-up after a PKB.
In conclusion, physicians must be aware of the risks of delayed bleeding potentially associated with adverse outcome, in patients undergoing PKB, especially in those with conventional risk factors for bleeding are present. Prolonged hospitalization for observation cannot be recommended as there are no additional risk factors in comparison with early bleed that can accurately predict patients of PKB at risk of delayed biopsy bleed, the variable time period to detect bleeding complication and low incidence of these complications. However, meticulous and more frequent outpatient monitoring after discharge and patient education on the signs and symptoms of potential complications after a PKB may be useful to detect this complication promptly and to intervene early to have good patient outcome.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
