This single-center, IRB-approved retrospective observational study included all adult patients (age ≥ 18) who underwent ultrasound guided transplant kidney biopsy over a 68-month period (January 3, 2017 through September 16, 2022), which totaled 1831 patients (1148 men, 681 women; age range 18–83 years; mean age, 50 years). The study was compliant with the Health Insurance Portability and Accountability Act. Informed consent was not required per institutional IRB and policies for this retrospective analysis.

Prior to procedure, serum laboratory values including complete blood count (CBC), prothrombin time (PT), partial thromboplastin time (PTT), international normalized ratio (INR) and basic metabolic panel (BMP) were routinely obtained. Patients with abnormal coagulation parameters such as elevated PT were corrected to 1.5 times normal with intravenous Vitamin K. Those with elevated PTT were corrected with fresh frozen plasma (FFP). Those with platelets less than 50 × 109/L received a platelet transfusion in accordance with Society of International Radiology (SIR) procedural guidelines [12]. Patients on anticoagulants withheld their medications according to the SIR management recommendations for high-risk bleeding procedures (i.e. Coumadin 5 days prior, aspirin 3–5 days prior) and were advised to re-start anticoagulation 12–48 h following biopsy (patient dependent) [12]. Patients with elevated serum blood urea nitrogen (BUN) > 20 were given intravenous DDAVP. Pre-procedure blood pressure goals were set at a systolic pressure less than 160 mm Hg and diastolic pressure below 90 mm Hg. For acute elevations in blood pressure, intravenous antihypertensive medications were administered. Biopsies were occasionally performed at higher pressures in patients with chronic blood pressure elevations per operator preference.

Most procedures were performed by a staff abdominal or interventional radiologist supervising a radiology fellow or resident in a dedicated procedure suite, with a minority of biopsies performed by the staff radiologist independently. In the supine position, all patients underwent either a full or abridged ultrasound scan of the transplant kidney with grayscale, color, and power Doppler evaluation of the renal parenchyma. Based on the ultrasound and additional scanning by the physicians, a suitable site was chosen for access to the upper or lower pole of the transplant kidney and the area cleansed with 1% chlorhexidine and draped with sterile blue towels and full drape of the lower extremities. For local anesthesia, 1% lidocaine was injected with a 25G needle subcutaneously and to the transplant renal capsule.

With ultrasound guidance, a 17G coaxial introducer needle was advanced to the transplant renal capsule. Depending on the attending radiologist preference, either an extracapsular or an intracapsular technique were used with either an 18G side-cutting biopsy needle (Mission or Max-Core, Bard, Murray Hill, NJ) or an 18G end-fire biopsy needle (Ultra, Biopince, Argon Medical, Plano, TX) was used (Fig. 2). The extracapsular technique consisted of guiding the 17G coaxial introducer needle to the renal transplant capsule with ultrasound guidance, followed by 1–2 punctures of the outer renal cortex with 18G biopsy device (typically Biopince Ultra). The intracapsular technique consisted of guiding the 17G needle approximately 5 mm into the renal cortex with ultrasound guidance, followed by 1–2 punctures of the outer renal cortex with 18G coaxial biopsy device (typically Bard Mission or Max-Core). This resulted in four total possible techniques: Intracapsular coaxial technique/side-cutting needle, intracapsular coaxial needle/end-cutting needle, extra-capsular coaxial technique/side-cutting needle, and extra-capsular coaxial technique/end-cutting needle. Regardless of which technique or biopsy needle was used, one tissue sample was obtained and analyzed by the on-site pathology technologist who evaluated the number and quality of glomeruli under light microscopy for diagnostic adequacy and determined if a second core was indicated. In most cases, two samples were obtained. Following biopsy, at the discretion of the performing attending, 1–3 gel-foam pledgets were loaded into a 1 cc syringe and injected through the 17G outer cannula. Pressure was applied above the transplant kidney for hemostasis. Finally, participants were monitored with ultrasound for up to 5 min to evaluate for immediate complications. They were then taken or back to a post-procedural area (outpatients) or returned to inpatient bed and instructed to remain supine for 3 h. No routine delayed renal ultrasound, urinalysis, or blood tests were performed unless there was symptomatic change or other clinical indication.

Side-cutting versus End-cutting needle. Schematic drawings illustrating the mechanism of two different core biopsy devices. (A) Traditional side-cutting biopsy device captures a semi-cylindrical core of tissue (i.e. Bard Mission and Bard Max-Core, CR Bard). (B) End-fire biopsy device captures a complete cylinder of tissue (i.e. Biopince, Argon medical)

Pre-procedure serum laboratory values, biopsy technique, biopsy device, number of cores, and sufficiency of biopsy sample were recorded. Patients were followed for bleeding complications up to one month after the procedure using SIR adverse event classification [13]. Major complications included patients requiring invasive intervention, either minor or major hospitalization, or permanent adverse sequelae. Minor complications included incidental findings attributed to the biopsy without clinical consequence, as well as patients requiring nominal therapy or overnight admission for monitoring.

Statistical analysis was performed using SPSS 28.0.1 for Windows (IBM, Armonk, New York). Fisher’s exact tests were performed to compare complication rates and biopsy sample sufficiency between the four sub-cohorts (extra-capsular technique with end-cutting needle, extra-capsular technique with side-cutting needle, intra-capsular technique with end-cutting needle, and intra-capsular technique with side-cutting needle). Additional Fischer’s exact tests were performed to compare complication rates and biopsy sample sufficiency between intra-capsular versus extra-capsular technique and the side-cutting versus end-cutting needle sub cohorts. Finally, comparison was made between patients with bleeding complications and patients without using a combination of independent t-test (Age, INR, Platelets) and Fischer’s exact test (whether tract embolization was performed). Two-tailed p-values less than 0.05 were considered statistically significant.