Osteogenesis Imperfecta, Perioperative Bleeding, and Desmopressin

A 72-yr-old man with adenocarcinoma of the prostate and transitional cell carcinoma of the bladder presented for cystoprostatectomy. The patient had a history of OI tarda that had manifested as multiple fractures in childhood and reduced hearing on the right side. His past medical history was also significant for hypertension and bleeding hemorrhoids. There was no history of spontaneous bleeding, easy bruising, or excessive surgical bleeding. The patient was not taking aspirin or nonsteroidal antiinflammatory drugs. Physical examination revealed the presence of blue sclera, a dorsal kyphosis, and bilateral reduction in hip mobility. His prostate gland was slightly enlarged and had areas of induration at the bases and right apex. His prostate-specific antigen level was 6.9 ng/ml (reference, < 5.6). Preoperative cystoscopic examination with biopsy revealed an invasive grade 3 transitional cell carcinoma of the bladder, and prostate biopsy results were positive for adenocarcinoma of the prostate. The cardiorespiratory examination was unremarkable. The patient's preoperative platelet count was 220 × 10⁹/l.

Standard American Society of Anesthesiologists monitors were placed. General anesthesia was induced with propofol and fentanyl and maintained with isoflurane in air and oxygen. An inline fluid warmer and Bair Hugger (Augustine Medical, Eden Prairie, MN) were used. Muscle relaxation was achieved using cisatracurium, and intermittent doses of morphine were used for analgesia. Marked vascularity and friability of the vessels complicated the dissection, leading to excessive, diffuse bleeding. Within 40 min of the incision, an estimated 1500 ml of blood had been lost, causing the patient's hemoglobin concentration to decrease from 13 g/dl (as measured preoperatively) to 8.7 g/dl. In light of the patient's hemodynamic instability and excessive, ongoing blood loss, 4 units of packed erythrocytes and 6 units of platelets were transfused. Despite this, the diffuse surgical bleeding continued, but no other bleeding sites were identified. Twenty minutes after administration of the platelets, 0.3 μg/kg (total dose, 22 μg) desmopressin acetate was administered in an effort to improve platelet function and to decrease blood loss. The drug was infused over 20 min to avoid severe hypotension. Within minutes of the infusion, the excessive, diffuse bleeding stopped. No further abnormal bleeding occurred, and the patient remained hemodynamically stable for the remainder of the procedure. In total, the patient received 6 units of packed erythrocytes, 6 units of platelets, 500 ml of 5% albumin, 500 ml hetastarch, and 5,600 ml lactated Ringer's solution. At the end of the case, the patient's hemoglobin concentration was 11.8 g/dl, and 4 h later his hemoglobin concentration was 11.2 g/dl, his platelet count was 199 × 10⁹/l, his prothrombin time was 11.9 s (reference range, 8.4–12.0 s), and his activated partial thromboplastin time was 25 s (reference range, 21–33 s). Prior to the onset of bleeding, the patient's temperature was 35.9°C, and it increased to 36.6°C after the bleeding episode. The patient's recovery and postoperative course were uneventful.

Osteogenesis imperfecta is a heterogenous group of inherited disorders of collagen type I caused by mutations of the COL1A1  or COL1A2  genes. 4Although the classic clinical description of OI is of a patient with brittle bones, blue sclera, and premature deafness, other organ systems are affected. The disease may cause cardiac valvular lesions, kidney stones, neurologic abnormalities, and metabolic derangements. In addition, the defect in collagen synthesis can cause excessive bruising and generalized oozing from wound sites. Surgical procedures performed in patients with OI carry a higher risk of bleeding complications. Laboratory investigations of patients with OI have demonstrated increased capillary fragility, decreased platelet retention, decreased levels of factor VIII, and deficient collagen-induced platelet aggregation. 5,6It is believed that the collagen defect results in friable tissues, small blood vessels that are unable to constrict adequately, and a defect in the platelet aggregation response around exposed subendothelium. The bleeding may occur despite normal results of coagulation studies and bleeding times, making predictions about intraoperative bleeding difficult. 7Coagulopathy with sudden development of widespread petechiae has been reported, 8as has fatal hemorrhage secondary to accidental rib fractures during spinal fusion surgery. 9 

The evaluation of platelet function and the role of platelet transfusion in OI are difficult to assess. Because OI is due to an impairment of the platelet-endothelial cell interaction, the thromboelastogram may not be useful. Indeed, we could not find a published case report regarding the use of the thromboelastogram in OI. The American Society of Anesthesiologists transfusion guidelines state, “platelet transfusion may be indicated despite an apparently adequate platelet count if there is known platelet dysfunction and microvascular bleeding.”10Bleeding time and platelet aggregation tests are not useful in the operating room, 10and “there is an urgent need for the development of clinically relevant measures of in vivo  platelet function and bleeding risk to guide the rational use of platelet transfusion.”11 

Desmopressin is a synthetic analog of antidiuretic hormone. It has a significant antidiuretic effect mediated via  V2 receptors and little or no vasoconstrictive effect (V1). 3,12The drug, mediated via  V2 effects, causes endothelial cells to release von Willebrand factor, tissue type plasminogen activator, and factor VIII:C. Desmopressin may be used in patients with uremia, chronic liver disease, and certain types of hemophilia, in which increased release of von Willebrand factor promotes platelet adhesiveness to the vascular endothelium and, hence, increases hemostatic activity. There are conflicting reports as to the utility of desmopressin in the reduction of blood loss during operations that involve substantial bleeding. Studies involving patients undergoing coronary artery bypass grafting, spinal fusion, and hip surgery have varied in their methodology and outcome. 13–16The adverse effects associated with the use of desmopressin are few. 3They include facial flushing, headache, hypotension, tachycardia, and the possibility of thrombosis. In addition, the antidiuretic effect may increase the difficulty of assessing the intravascular volume status. There have been a few reports of hyponatremia and seizures, mostly in children.

Desmopressin works in congenital and acquired platelet disorders by increasing plasma concentrations of factor VIII, von Willebrand factor, and tissue plasminogen activator and by increasing platelet adhesiveness. 3,12The hemostatic effect is almost immediate after administering an intravenous dose of 0.3 μg/kg. 3Desmopressin may be useful for improving hemostasis and ultimately decreasing transfusion requirements in the surgical patient with OI. By increasing platelet deposition onto vascular subendothelium, desmopressin would seem to be an excellent adjunct to the perioperative management of patients with OI.

We cannot say definitively that desmopressin stopped the bleeding in this case, but circumstantial evidence supports this hypothesis. A review of the literature failed to reveal any previous reports of the use of desmopressin to treat the coagulopathy associated with OI. Although further investigation is warranted, due to the low incidence of OI and the difficulty in documenting the coagulopathy associated with OI, randomized, controlled trials will be difficult to perform. Qualitative tests of platelet function, performed before and after the administration of desmopressin, may help to elucidate the drug's role in this setting. However, given the relatively benign side effect profile of desmopressin and the theoretical basis for its effectiveness, consideration should be given to the use of desmopressin in patients with OI undergoing major surgery when complicated by significant microvascular bleeding.