Cardiovascular Instability Requiring Treatment after Intravenous Heparin for Cardiopulmonary Bypass

We have observed hypotension (mean arterial pressure [MAP] < 60 mm Hg) and/or bradycardia (heart rate [HR] < 40 bpm) shortly after IV heparin loading for cardiopulmonary bypass (CPB) in several patients undergoing cardiac surgery. Serum electrolytes after these episodes of cardiovascular instability (CVI, defined as MAP < 60 mm Hg and/or HR < 40 bpm) often showed relative hyperkalemia. Hyperkalemia has been reported with chronic heparin administration (1–6), but only after several days of treatment, and has not been reported to occur acutely. The purposes of this report were to review the characteristics of the cases of CVI and to compare the cases of CVI with a control group. Methods Records of all patients undergoing cardiac surgery requiring CPB during a 6-mo period were reviewed to identify all episodes of CVI occurring within 10 min of IV heparin loading. Two hundred fifty-six patient records were reviewed. CVI occurred in 16 patients. The remaining 240 patients were considered the control population. All patients had blood analyzed in a similar fashion intraoperatively. The first sample was obtained from the arterial catheter, immediately after the anesthetic induction, for measurement of pH, PCO2, and hematocrit, as well as serum potassium (K+), sodium (Na+), and calcium levels. Porcine heparin (400 IU/kg) was administered to achieve an activated clotting time of ≥400 s. Another blood sample was obtained 5 min after the heparin bolus was administered. Analysis was performed on all blood using the same equipment, which was calibrated daily. Data recorded included patients’ age, weight, ejection fraction, serum creatinine, sex, history (and type) of diabetes, and history of use of angiotensin-converting-enzyme inhibitors, calcium channel blockers, beta-blockers, and any preoperative heparin infusion. Mean HR and MAP immediately preheparin and at the time of blood sampling were recorded. Statistical analysis was performed on continuous data using the paired or unpaired Student’s t-test as appropriate, with the Bonferroni correction applied as necessary. Categorical data were compared using the χ2 or Fisher’s exact test. Results CVI occurred in 16 of the 256 patients (6.25%). In the 16 patients with CVI, MAP decreased from 73.6 ± 14.3 mm Hg preheparin bolus to 56.7 ± 7.4 mm Hg postbolus, as mean HR decreased from 70.6 ± 9.3 bpm to 56.7 ± 12.1 bpm. Because the CVI was always treated, it is unknown if it would have continued or if spontaneous recovery might have occurred. All episodes of hypotension were initially treated with 5-mg increments of ephedrine IV. Thirteen patients had restoration of MAP to 60 mm Hg after 10 mg of ephedrine. Three patients also required 100 μg of neosynephrine. Two of the latter patients received CPB urgently because of sustained hypotension. The reductions in HR and MAP were highly significant statistically (P = 0.0001 and P = 0.0004, respectively). They correlated positively with each other and negatively with the change in potassium level, which increased significantly (P = 0.0001, Table 1).Table 1: Case-Control ComparisonComparison of the cases and controls (Table 1) showed that the cases were distinguished from the control group primarily by a much larger increase in serum K+ after IV heparin administration (1.94 vs 0.50 mmol/L). The final serum Na+ was significantly lower among cases than controls (134.5 vs 137.3 mmol/L). Beta-blockade was used significantly more often among controls (P = 0.04, Table 1), and type I diabetes tended to be more common among cases (P = 0.08, Table 1). Other factors potentially affecting serum K+, such as use of angiotensin-converting-enzyme inhibitors, heparin infusion, presence of renal insufficiency, and ejection fraction, were not significantly different between groups (Table 1). Discussion The principal findings of this report are that CVI occurred shortly after IV heparin administration for CPB in 16 out of 256 patients (6.25%) and that serum K+ increased by 1.94 mmol/L in patients with CVI, while serum K+ increased by a mean of 0.50 mmol/L in control patients without CVI. CVI was arbitrarily defined as MAP < 60 mm Hg and/or HR < 40 bpm. MAP and HR may have been reduced in other patients, but to a lesser degree; however, this could not be determined accurately from the available data. Further description of the hemodynamic changes associated with bolus IV heparin administration for CPB would be appropriate for prospective evaluation and is in progress. In the treated patients, hypotension, bradycardia, and the increase in K+ were closely associated. However, the classic electrocardiographic signs of hyperkalemia were not seen in any of the treated patients. The cases and controls could not be differentiated on the basis of arterial PCO2 or pH (Table 1) (7). The mean serum Na+ decreased in patients with CVI, suggesting that slight volume expansion actually occurred. It is unlikely that the observed increase in serum K+ was caused by a ventilation or volume change. Chronic administration of heparin (for days), has been associated with an increase of serum K+ of about 0.5 mmol/L (1–4). Acute administration of heparin in vitro has caused slight elevations of serum K+ (8), but IV heparin administration has not been noted to affect serum K+ acutely in humans. None of the usual medications or disease processes implicated in increasing serum K+ were found to be significant in this study. Indeed, a trend to an attenuated rise in serum K+ in patients on beta-blockers was identified. In summary, acute IV heparin (400 IU/kg) administration was closely associated with CVI in 16 of 256 patients (6.25%) undergoing CPB. A strong association between CVI and an increase in serum K+ was identified in the treated cases. Mean serum K+ rose modestly in control patients without CVI, by 0.5 mmol/L, but it increased by 1.94 mmol/L in treated cases. The mechanism of this increase in serum K+ after acute IV heparin administration is unknown, and further investigation is warranted.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.