Current complications of diagnostic and therapeutic cardiac catheterization.

Data from 2,883 cardiac catheterizations performed during an 18 month period (from July 1986 through December 1987) were analyzed to assess the current complication profile of diagnostic and therapeutic procedures. Procedures performed during the study period included 1,609 diagnostic catheterizations, 933 percutaneous transluminal coronary angioplasties and 199 percutaneous balloon valvuloplasties. Overall, the mortality rate was 0.28% but ranged from 0.12% for diagnostic catheterizations to 0.3% for coronary angioplasty and 1.5% for balloon valvuloplasty. Emergency cardiac surgery was required in 12 angioplasty patients (1.2%). Cardiac perforation occurred in seven patients (0.2%), of whom six were undergoing valvuloplasty, and five (2.5% of valvuloplasty attempts) required emergency surgery for correction. Local vascular complications requiring operative repair occurred in 1.9% of patients overall, ranging from 1.6% for diagnostic catheterization to 1.5% for angioplasty and 7.5% for valvuloplasty. Although the complication rates for diagnostic catheterization compare favorably with those of previous multicenter registries, current overall complication rates are significantly higher because of the performance of therapeutic procedures with greater intrinsic risk and the inclusion of increasingly aged and acutely ill or unstable patients.

Incidence and management of limb ischemia with percutaneous wire-guided intraaortic balloon catheters.

In 103 patients who underwent placement of 106 percutaneous wire-guided intraaortic balloon catheters between August 1983 and January 1986, all placements were successful and the average duration of counterpulsation was 3.4 +/- 1.6 days. During counterpulsation, 45 patients developed limb ischemia that required premature balloon removal in 29 patients. The development of limb ischemia was significantly related to the presence of diabetes (risk ratio 2.0), peripheral vascular disease (risk ratio 1.9), female gender (risk ratio 1.8) and the presence of a postinsertion ankle-brachial pressure index less than 0.8 (risk ratio 7.9). There was no association between the development of limb ischemia and age, body surface area, balloon size (10.5F/12F) or adequacy of anticoagulation. Fifteen patients underwent vascular surgery for treatment of balloon-related limb ischemia, which was associated with one operative death. Nine patients had persistent limb ischemia (seven asymptomatic, two symptomatic) at the time of hospital discharge. Improvements in wire-guided balloon technology have increased the probability of successful balloon placement over that of surgical placement and have reduced the incidence of major aortic injury, but there is no evidence that these improvements have reduced the incidence of limb ischemia or its sequelae. This should be borne in mind before proceeding with balloon insertion in patients with one or more risk factors for developing limb ischemia.

Is lower extremity revascularization worthwhile in patients with end-stage renal disease?

BACKGROUND: The purpose of this study was to review the results of lower extremity revascularization in patients with end-stage renal disease and to determine in these patients the functional benefit and cost of an aggressive approach to limb preservation. METHODS: During a 5-year period at our institution, 33 bypass operations were performed on 31 limbs of 23 dialysis-dependent patients. Indications for revascularization were limited (18) or extensive (12) tissue loss or ischemia without tissue loss (3). Procedures included aortobifemoral bypass (1), femoropopliteal bypass (10), and femorotibial/pedal bypass (22). A digital or transmetatarsal amputation was performed in 57% of limbs. RESULTS: The 30-day primary patency was 100%. Cumulative primary and secondary patency rates at 2 years were 65% and 79%, respectively. Limb salvage was 67% and 59% at 1 and 2 years, respectively. Patient survival was poor (47% at 2 years). Peritoneal dialysis was predictive of poor survival (P <.001). Four of 5 patients on peritoneal dialysis died within 3 months of intervention. Extensive tissue loss was predictive of a diminished rate of limb salvage (P =.027). Only 39% of limbs with extensive tissue loss were salvaged at 1 year compared with 78% and 100% of limbs with limited and no tissue loss, respectively. The average hospital cost was $44,308 per year of limb salvage. CONCLUSIONS: Although revascularization of ischemic limbs in dialysis patients can be achieved with an excellent initial graft patency and reasonable limb salvage, patient survival is poor and costs are high. A selective approach to revascularization in these complicated patients may be indicated. For patients treated with peritoneal dialysis and for those with extensive tissue loss, primary amputation may be the preferred approach.

Platelet attachment stimulates endothelial cell regeneration after arterial injury.

BACKGROUND: Arterial injury is associated with endothelial disruption and attachment of platelets to an exposed subintimal layer. A variety of factors released by platelets may affect the ability of endothelial cells bordering an injury to regenerate. In this study an organ culture model of arterial injury was used to investigate the relationship between attachment of platelets to a superficial arterial injury and endothelial regeneration. METHODS: A defined superficial endothelial injury was made in whole vessel wall explants of rabbit thoracic aorta. Injured explants were treated with either fresh whole platelets, the supernatant of platelets aggregated by collagen, or basic fibroblast growth factor. Four days after injury and treatment, the average distance of endothelial regeneration was determined. RESULTS: A dramatic increase in the rate of endothelial cell regeneration was observed when injured vessels were exposed to fresh whole platelets (p = 0.003). This increase in regeneration was comparable to that observed with fibroblast growth factor. No increase in the regenerative rate was found after exposure of explants to the supernatant of aggregated platelets (p = 0.69). CONCLUSIONS: Platelets stimulate endothelial regeneration at a rate equal to that observed with the potent endothelial mitogen basic fibroblast growth factor. Because this effect was not demonstrated with the supernatant of aggregated platelets, endothelial regeneration may be dependent on attachment of the platelets to the area of injury.

Randomized trial of albumin vs. electrolyte solutions during abdominal aortic operations.

In a prospective randomized trial of 16 patients undergoing abdominal vascular reconstructive procedures, changes in plasma volume, serum oncotic pressure (pis), serum albumin and total protein concentration, alveolar to arterial oxygen tension differences (AaDO2, FIO2 = 1.0), creatinine clearance, body weight, and fluid and sodium intake were examined. By random assignment patients received either an albumin- or a sodium-rich intraoperative fluid regimen. Pulmonary arteriovenous admixture was significantly less in the albumin group (n = 7) than in the electrolyte group (n = 9) on the first postoperative day. The change in AaDO2 correlated positively with the total sodium intake in the electrolyte group. Despite the larger fluid load and significantly greater gain of body weight, patients in the electrolyte group had a postoperative plasma volume significantly lower than the preoperative value. Postoperative values of albumin concentration, circulating albumin mass and pis were significantly greater in the albumin group in comparison to the electrolyte group. Creatinine clearance values were not different between the two groups. The change in pis correlated significantly with sodium intake and circulating albumin mass. Pulmonary shunting and expansion of the extracellular fluid volume may be minimized without adverse effects on renal function by administration of fluids rich in albumin in preference to sodium.

The effect of carbenoxolone on changes in canine and human gastric mucosa caused by taurocholate and ethanol.

The effect of carbenoxolone on taurocholate-induced changes in gastric mucosal permeabiity was assessed in three dogs, each of which was prepared with an antrectomy and a vagally denervated gastric pouch. Perfusion of the canine pouches with sodium taurocholate (40 mM) was associated with an increase in hydrogen ion back diffusion. This effect was not diminished by 10 days of carbenoxolone treatment. The effect of carbenoxolone on ethanol-induced changes in gastric mucosal permeability was assessed in six normal human subjects. A significant increase of gastric mucosal permeability was observed in six normal human subjects after instillation of ethanol (20 percent v/v). After 3 weeks of oral ingestion of carbenoxolone, there was inconsistent protection against ethanol-induced increases in gastric mucosal permeability. Basal secretion of hydrogen ion and postethanol hydrogen ion secretion appear to be diminished by carbenoxolone. These studies suggest that carbenoxolone does not protect against taurocholate- and ethanol-induced increases in gastric mucosal permeability in the dog and in man. It seems unlikely that carbenoxolone exerts its beneficial effect on the healing of gastric ulcers in man by an effect on gastric mucosal permeability.

Cardiorespiratory and sympathoadrenal responses during weaning from controlled ventilation.

Weaning from controlled ventilation was attempted in 20 patients who were ventilated with volume constant ventilators for at least 24 hours before study. Measurement of alveolar to arterial oxygen tension differences on 100% oxygen (Aa DO2 1.0) in patients who failed to wean (nonweaners, n = 10), showed a mean AaDO21.0 at 5 minutes after beginning to wean of 388 +/- 56 mm Hg (SEM), and right-to-left shunt (QS/Qt) of 21% values which were significantly different from control (p less than 0.025) and significantly different from ten patients who weaned (p less than 0.005). Since rises in QS/Qt were not accompanied by increased pulmonary capillary wedge pressure, the increased QS/Qt was most likely due to acute atelectasis and not left ventricular failure. These data suggest that patients who fail an initial period of weaning should be placed on positive end-expiratory pressure during subsequent weaning attempts. Significant increases in cardiac output and atrial PCO2 occurred in both groups. Assessment of urinary catecholamine excretion during weaning suggests that sympathoadrenal stimulation often is intense and usually is greater in those patients who weaned successfully. Increased arterial PCO2 may be the mechanism for sympathoadrenal stimulation and rises in cardiac output (r = 0.39, p less than 0.01) during weaning.

Thromboembolism after insertion of the Mobin-Uddin caval filter.

The occurrence, in one patient, of a life-threatening pulmonary embolus 3 days following insertion of a Mobin-Uddin umbrella filter promoted a review of our experience with this device. Over the past 5 years, 41 patients had umbrella filters inserted at the Beth Israel Hospital. Five patients (12%) had proven or probable pulmonary embolization following filer placement. In four of the seven autopsied patients, thrombus was found on the cardiac side of the umbrella, and in one of these patients a fresh embolus was also found in the pulmonary artery. We conclude that the standard (nonheparin-impregnated) Mobin-Uddin umbrella device offers incomplete protection against pumonary embolization in patients who cannot concurrently receive anticoagulant drugs.

Prospective study of wound complications in continuous infrainguinal incisions after lower limb arterial reconstruction: incidence, risk factors, and cost.

BACKGROUND: Wound complications after lower extremity arterial reconstruction can range from a minor lymphatic leak that causes minimal disability to a severe infection that jeopardizes the limb and life of the affected patient. This study was designed to define more clearly the incidence, severity, and the cost of these complications. METHODS: During a 1-year period the infrainguinal incisions of all patients undergoing lower limb arterial reconstruction were evaluated prospectively. One hundred fifty-six infrainguinal incisions were monitored serially for the presence of infection, hematoma, seroma, serous leak, necrosis, or wound dehiscence. The need for additional treatment or services related to these complications and the cost of these services were determined. RESULTS: Complications occurred in 10% of 77 infrainguinal incisions that were isolated to the groin (groin incisions) e.g., after aortobifemoral bypass, femoral endarterectomy). In only one of these patients was significant cost related to treatment of a complication. Complications occurred in 44% of 79 incisions used for femoral popliteal/tibial and pedal bypasses (distal incisions). In this latter group independent predictors of any complication were age (p=0.02) and obesity (p=0.05); predictors of in-hospital infection were preoperative evidence of venous stasis (p=0.01) and preoperative infection in the same extremity (p=0.08). Fifteen distal wound complications provided additional expense related to reoperation, extended hospitalization or rehospitalization, and rehabilitation or visiting nurse services, with a mean cost per patient undergoing reconstruction of $688. CONCLUSIONS: After lower limb arterial reconstruction, infrainguinal wound complications in isolated groin incisions produce minimal morbidity and cost, whereas complications in incisions after distal bypass are both frequent and associated with significant additional expense.

Prevention of deep vein thrombosis in urological patients: a controlled, randomized trial of low-dose heparin and external pneumatic compression boots.

A prospective, randomized clinical trial in 83 patients undergoing open urological operations was performed to compare the relative prophylactic efficacy of low-dose heparin and external pneumatic compression (EPC) of the calves in comparison to no treatment in the prevention of deep vein thrombosis (DVT). The diagnosis of deep vein thrombosis was established by the I125 fibrinogen scan and contrast phlebography. The data indicate that EPC is more effective than no treatment (p less than 0.04) and more effective than low-dose heparin (p less than 0.04) in the prevention of DVT. There was one pulmonary embolus in each of the three groups, and in the treatment groups this occurred without evidence of DVT.

Albumin synthesis and nitrogen balance in postoperative patients.

In a continuing prospective randomized study in a series of patients in the postoperative state, the rate of albumin synthesis was measured using the 14C carbonate technique. The majority of the patients studied were subjected to partial or total colectomy, and each patient was randomized to receive a hypocaloric intravenous regimen consisting of 3.5% amino acids (essential and nonessential) or 3.5% amino acids with 2.5% glucose. A mean (+/- SEM) of 75.7 +/- 2.7 gm/day of amino acids was given to the seven patients in the amino acid group. A mean of 74.0 +/- 5.9 gm/day of amino acids and 52.8 +/- 4.2 gm/day of glucose was infused in the group of five patients given amino acids with glucose. The albumin synthetic rate (mg/kg/day) measured on the fourth day after operation in the amino acid alone group was 233.9 +/- 27.9 as compared to 204.1 +/- 24.4 in the group given amino acids with glucose (P = NS). The urea synthesis rates (mg/kg/day) in the two groups were 334.2 +/- 47.8 and 250.3 +/- 27.5, respectively, and the net Kjeldahl nitrogen balances (grams per day) were -4.98 +/- 1.92 and -2.68 +/- 1.49, respectively. We conclude that infused amino acids promote the synthesis of albumin equally effectively in the presence or absence of a small amount of added glucose. Glucose may reduce slightly the pool of amino nitrogen available for albumin and urea synthesis by diverting some of the infused amino acids from protein synthesis by the liver to muscle.

Effect of isocaloric fat or glucose on albumin synthesis and nitrogen balance in patients receiving amino acid infusion.

Twelve patients who had undergone colon operations were randomized prospectively in the postoperative recovery room. Seven received a hypocaloric intravenous solution consisting of 3.5% amino acids plus 2.5% glucose and five received 3.5% amino acids plus 10% fat. Nitrogen balance data indicated that isocaloric amounts of glucose and fat seem equally effective when combined with amino acids. Although the albumin synthesis rates (measured by the 14C technique on the fourth postoperative day) were not significantly different between the two groups, when the results were included with pooled data from previous similar studies, there was a statistically significant difference which indicated a higher rate of albumin synthesis in patients who received the combination of amino acids and fat compared with those who received amino acids and glucose P < 0.05). In this clinical setting, fat may favor the uptake and synthesis of amino acids into visceral proteins, while glucose may have a more direct role in the synthesis of skeletal muscle protein.

Prevention of deep vein thrombosis in neurosurgical patients: a controlled, randomized trial of external pneumatic compression boots.

A prospective, randomized clinical trial in 95 patients undergoing neurosurgical operative procedures was performed to investigate the efficacy of external pneumatic compression (EPC) of the calves as compared with results in a control group that received no specific form of prophylaxis for prevention of deep vein thrombosis (DVT). The diagnosis of DVT was established by the I125 fibrinogen scan and radiographic contrast phlebography. The data indicate that EPC provides significant protection against the development of DVT in comparison with results in the control group (p less than 0.005). There were no known pulmonary emboli in any of the EPC-treated patients. There were no complications of EPC.

Energy intake can determine albumin synthesis in man after surgery.

Albumin synthesis rate, nitrogen balance, plasma hormone levels, and selected substrates were measured after operation in 12 patients who underwent colonic operations who were randomized to receive an intravenous fluid regimen that contained either 3.5% amino acids with 20% fat and 2.5% glucose or 3.5% amino acids with 20% fat alone. The albumin synthesis rate was higher in patients who received the first of these intravenous mixtures (357 +/- 34 mg/kg/day versus 216 +/- 22 mg/kg/day; p less than 0.01), but they also had a significantly higher intake of calories (10.2 +/- 1.1 calorie/kg/day versus 6.4 +/- 0.6 calorie/kg/day; p = 0.01). The mean albumin synthesis rate in the group who received amino acids with glucose and fat is the highest we have measured in our series of studies. Although a previous trial in a similar group of patients suggested that glucose acts on albumin synthesis by diverting uptake of amino acids into skeletal muscle, it is possible that the impressive increase in the albumin synthesis rate in patients of the present series who received supplementary glucose is related to the extra energy infused. In contrast, nitrogen balance was similar in both groups, and thus was not predictive of protein synthesis. In addition, myofibrillar protein degradation appears to be equivalent in the two groups, as indicated by 3-methylhistidine output. Plasma albumin synthesis thus may be sensitive, especially to energy intake.

Do neck incisions influence nerve deficits after carotid endarterectomy?

OBJECTIVES: To determine whether transverse neck incisions for carotid endarterectomy were associated with a similar or greater incidence of cranial nerve complications when compared with vertical skin incisions, and to assess the patient's perception of the appearance of the incision. DESIGN: Prospective, but not randomized. SETTING: A university-affiliated tertiary care hospital. PATIENTS/INTERVENTIONS: Eighty-five consecutive carotid endarterectomy procedures were evaluated prospectively in 80 patients. Although patients were not randomly assigned, consideration was given to having approximately the same number of patients who had carotid endarterectomy performed through transverse neck incision as through vertical neck incision. Forty-four carotid endarterectomies were performed with a vertical incision and 41 procedures were performed with a transverse incision. MAIN OUTCOME MEASURE: To determine the incidence of cranial nerve dysfunction (primarily nerves VII and XII) after operation. RESULTS: The incidence of palsies of cranial nerves VII and XII in the two groups was similar; there was no statistical significance (the seventh nerve palsy, 32% transverse vs 25% vertical; the 12th nerve palsy, 15% transverse vs 20% vertical). Seventy-two percent of the deficits had disappeared by the 3- to 6-month follow-up. Patients expressed a clear preference for the transverse incision (P = .04). CONCLUSIONS: Although surgical exposure was simpler with the vertical incision, adequate exposure with the transverse incision was always possible. The incidence of mostly temporary deficits of cranial nerves VII and XII was similar. Patients favored the transverse incision.

Vascular complications of percutaneous femoral cardiac interventions. Incidence and operative repair.

Of 7333 patients undergoing percutaneous left-heart catheterization procedures during a seven-year seven-month period, 73 patients (1%) underwent 75 operative repairs of catheterization-related vascular complications. The overall incidence of operative repair varied according to the type of percutaneous femoral artery procedure performed: 0.6% for diagnostic heart catheterization, 0.9% for percutaneous transluminal angioplasty, 5.2% for transfemoral balloon valvuloplasty, and 11.5% for intra-aortic balloon pump placement. This suggests that (1) femoral and iliac artery occlusions can and should be repaired promptly while the patient is under local anesthesia; (2) false aneurysms in this clinical setting can be approached directly through the aneurysm cavity; (3) the diagnosis of false aneurysm or arteriovenous fistula can usually be made on clinical grounds alone, without resorting to angiography; (4) venous or arterial patch angioplasty is the preferred technique for small or severely traumatized femoral arteries; and (5) the necessity for operative repair of these lesions will continue to increase in frequency as percutaneous cardiologic diagnostic and therapeutic interventions are used more widely.

Cardiac performance testing by volume loading and lower-extremity compression.

Cardiac performance was tested with a combination of volume loading and lower-extremity compression by an anti-g suit in 11 patients scheduled for aortic operations. The mean response of cardiac output (CO) to volume loading and anti-g suit compression was flat, despite a significant rise in pulmonary capillary wedge pressure (PCWP). Right to left intrapulmonary shunt was unchanged by the testing procedure. The low mean wedge pressure reemphasizes the need for preoperative insertion of a Swan-Ganz catheter and for measurement of PCWP and Co to provide optimal fluid treatment to each patient who undergoes an aortic reconstructive operation.

Pulmonary management of the urologic patient.

Morbidity and mortality from pulmonary complications following urinary tract operations can be reduced by preoperative identification of the high risk patient. Pulmonary function tests and arterial blood gases are necessary to identify these patients and to delineate the severity of their pulmonary disease. Respiratory complications can be prevented in many patients with the proper use of pre- and postoperative chest physical therapy and oxygen therapy. Despite the most careful pulmonary management, some patients develop acute respiratory failure following urologic operations. Respiratory failure results from a combination of physiologic abnormalities which impair alveolar ventilation and oxygenation. Utilizing controlled ventilation, supplemental oxygen, and a physiologic approach to treating the underlying cause of respiratory failure, three fourths of urologic patients in respiratory failure may be expected to survive.

15 (R)-15-methyl prostaglandin E2 does not prevent gastrointestinal bleeding in seriously ill patients.

A prospective, randomized trial was designed to compare the relative efficacy of 15 (R)-15-methyl prostaglandin E2 with antacid (usually Mylanta II) in 46 patients admitted to a respiratory-surgical intensive care unit. Bleeding was assessed by a modification of the Hemoccult slide test. Three of 22 patients in the antacid group bled, and 12 of 24 patients in the prostaglandin group bled, for a highly significant difference (p = 0.008). Patients in whom prophylaxis failed tended to have a greater number of risk factors. Other prostaglandin analogues that do not require conversion from an inactive to an active form, may be more useful than the agent we studied. Based on currently available data, the hourly titration of the gastric juice to a pH of greater than 3.5 remains the preferred method of prophylaxis for acute bleeding from the stomach in seriously ill patients.

Improving cost efficiency on a vascular surgery service.

BACKGROUND: A vascular task force (VTF) consisting of two vascular surgeons and other key personnel was established to reduce costs and improve efficiency in the management of patients on a vascular surgery service. METHODS: The VTF met monthly beginning in 1994 to study and implement changes in the management of patients with (1) abdominal vascular, (2), carotid endarterectomy (3) distal bypass, and (4) other vascular procedures, including amputations. Length of stay, and fixed and variable costs were assessed for change over time using Pearson correlation coefficients. RESULTS: Improvements in efficiency (length of stay) and decreases in costs (fixed and variable costs) from fiscal year 1993 to fiscal year 1996 were significant for the total group of vascular patients (P