Desmopressin and Bleeding after Kidney Biopsy

Desmopressin before kidney biopsy: Which patients do (and don’t) benefit?

Athavale is a nephrologist and director of clinical research at Cook County Health. His coauthors are Hemant Kulkarni, _MD, of M&H Research, San Antonio, Texas; and Cagil D. Arslan, MD, and Peter Hart, MD, of Cook County Health.

Their study included 269 patients who underwent percutaneous kidney biopsy at the authors’ urban public hospital from 2014 through 2018. All patients had available data on bleeding time; patients with bleeding time over 10 minutes, platelet counts under 50,000, or evidence of coagulopathy were excluded from the analysis. Indications for biopsy (nonexclusive) included nephritis in 122 patients, nephrotic syndrome in 122, and chronic kidney disease in 85. Biopsy was performed on an emergency basis in 56 patients.

At the discretion of the nephrologist performing the biopsy, 37.17% of patients received desmopressin (0.3 µg/kg IV). There were some significant differences in patient characteristics, including potential risk factors for bleeding events: Patients receiving desmopressin had lower baseline hemoglobin, lower platelet count, lower estimated glomerular filtration rate, higher bleeding time, higher blood urea nitrogen, and higher serum creatinine.

Athavale and colleagues looked at whether desmopressin achieved the goal of reducing bleeding risk, and whether this effect differed for patients with decreased versus normal kidney function. The primary outcome was a composite of a 1 g/dL or greater decrease in hemoglobin, gross hematuria, and need for angiogram or red blood cell transfusion. Overall, patients in the desmopressin group had a higher rate of bleeding events: 59.46%, compared to 31.75% in those who did not receive desmopressin.

A propensity score was generated to account for variables that differed between groups. In this analysis, the odds of postbiopsy bleeding were nearly four times higher in patients receiving desmopressin: odds ratio 3.88. Most of the difference was related to higher rates of decreased hemoglobin (44.14% versus 18.96%) and hematoma (18.02% versus 15.17%) in desmopressin-treated patients. Rates of gross hematuria and need for red blood cell transfusion were similar between groups.

On subgroup analysis, two factors contributed to the desmopressin-related increase in bleeding events: high baseline eGFR and low serum creatinine. None of the other factors analyzed—including gender, emergent versus elective biopsy, acute kidney injury, diabetes, hypertension, or bleeding time—were significantly related to bleeding risk.

For patients with high baseline creatinine (1.8 mg/dL or greater), there was a trend toward reduced bleeding complications after desmopressin administration: OR 2.11 (95% confidence interval 0.87 to 5.11). In contrast, administration of desmopressin to patients with low baseline creatinine (less than 1.8 mg/dL) was associated with a large increase in bleeding risk: OR 9.2 [95% confidence interval 2.95 to 31.96].

“This increased odds of bleeding was driven mainly by a drop in hemoglobin in patients with relatively preserved kidney function,” Athavale said. “Because these patients did not need blood transfusion or angiographic embolization and the absolute magnitude of hemoglobin drop was small, we felt that this decrease in hemoglobin reflected dilution effect after desmopressin administration and not true bleeding from kidney tissue.”

New questions on routine desmopressin before kidney biopsy

Bleeding is the most frequent complication of percutaneous kidney biopsy and is more common in patients with decreased kidney function. Several abnormalities may contribute to platelet dysfunction in patients with kidney disease. The rationale for using desmopressin is to improve platelet aggregation by increasing release of von Willebrand factor.

“However, there is sparse data to support routine use of desmopressin, and it is not known if there is a subset of patients who will benefit the most from administration of desmopressin,” Athavale and coauthors write.

Some studies have reported lower bleeding risk in patients receiving desmopressin. In a 2004 study, Manno et al. reported a significant reduction in postbiopsy bleeding with desmopressin: 13.7% versus 30.5%, relative risk 0.45 (2). However, the only benefit was in the number and size of hematomas, with no significant effect on hard endpoints such as transfusion and angiographic intervention.

In a registry study, Peters et al. found a lower overall rate of complications (3.4% versus 8.4%, odds ratio 0.39) with desmopressin before native kidney biopsy in patients with serum creatinine over 150 µmol/L, particularly in women (3).

However, other studies have yielded conflicting results. A recent retrospective study by Lim et al. found no significant difference in bleeding risk for patients who did and did not receive desmopressin (15.0% vs. 13.3%) (4). However, these authors observed a sharply increased risk of severe hyponatremia in patients receiving desmopressin: 10.7% versus 3.0%, adjusted odds ratio 4.02. Athavale and colleagues found no episodes of symptomatic hyponatremia, nor any other adverse events or side effects attributable to desmopressin.

In a study in CJASN, Moledina et al. found an 8% rate of transfusion, a 7% rate of hematoma, and a 2% rate of angiographic intervention in a cohort of 256 patients undergoing kidney biopsy (5). Hospitalized patients were at higher risk of complications; other risk factors included lower platelet count, female sex, and higher blood urea nitrogen (BUN). In this analysis, desmopressin was associated with a lower risk of transfusions after controlling for BUN level: odds ratio 0.24.

The new study suggests that desmopressin reduces postbiopsy bleeding in patients with elevated serum creatinine (1.8 mg/dL) or higher but is not useful in patients with normal creatinine levels. Athavale and colleagues note that most of the desmopressin-related increase was driven by an increase in postbiopsy hemoglobin, which doesn’t necessarily reflect true bleeding from the kidney. However, a drop in hemoglobin may lead to additional and unnecessary testing, leading to increased costs and patient anxiety. The authors add that desmopressin has been linked to other adverse events, including increased thrombotic risk when given to reduce bleeding risk in non-uremic patients undergoing major cardiovascular surgery.

Athavale and colleagues note some strengths of their study, including the inclusion of many patients with common risk factors: elevated serum creatinine in about 70%, high body mass index in 40%, and acute kidney injury in 4%. The experience reflects contemporary kidney biopsy practice. All procedures were performed using an 18-gauge biopsy needle; the findings may not be applicable to biopsies performed using a 16-gauge needle.

These limitations, together with the mixed findings and sparse evidence on bleeding risk associated with kidney biopsy in general and the effects of desmopressin in particular, highlight the need for definitive studies of assessing and reducing bleeding risk. Athavale and coauthors conclude, “A randomized trial is needed to further evaluate the efficacy and safety of desmopressin in high-risk patients.”