ADVERTISEMENT
Original Contribution
In-Hospital Complications of Peripheral Vascular Interventions Using Unfractionated Heparin as the Primary Anticoagulant
May 2003
Unfractionated heparin is the current antithrombotic of choice in peripheral vascular interventions. Heparin has an unpredictable anticoagulation response, is an indirect thrombin inhibitor, does not inhibit bound thrombin and activates platelets. A contemporary in-hospital complication rate following percutaneous transluminal peripheral angioplasty (PTA) has not been well defined with the use of unfractionated heparin as the initial primary anticoagulant. In-hospital complication rates following PTA have been reported in the literature; they range from 3.5% to as high as 32.7%.1–7 These studies differed in their inclusion and exclusion criteria, designs, and patient numbers and characteristics, and undertook no consistent reporting on anticoagulation regimen utilized. Given the continuous search for a better anticoagulant than heparin during percutaneous interventions, it is important to define our current baseline complication rate during PTA with heparin as the primary anticoagulant.
In this single-center experience, we report on our complication rate during PTA in 131 consecutive patients who received heparin as the primary anticoagulant, and model the primary adverse event of salvage revascularization and any major complications.
Methods
Two-hundred and thirteen patients who underwent PTA (all except carotid angioplasty) from November 1, 2000 to November 1, 2001 at our institution were consecutively enrolled in this study. Of these, eighty-two were excluded because of: 1) two or more planned staged peripheral procedures during the same hospital stay; 2) acute myocardial infarction preceding PTA; 3) the use of elective adjunctive intravenous glycoprotein IIb/IIIa inhibitors during the procedure; 4) concomitant coronary procedures; or 5) continuous heparin drip prior to the procedure. All charts were retrospectively reviewed for clinical, angiographic and serious event rates. An interventional cardiologist not involved in the procedure adjudicated the in-hospital serious adverse events (SAE). SAE were defined as follows: 1) major bleed, defined as requiring >= 2 units of packed red blood cell transfusion, retroperitoneal bleed, or a drop of hemoglobin (Hb) after the procedure by more than 3 g/dl; 2) vascular complications, defined as arteriovenous fistula or pseudoaneurysm after the procedure when suspected clinically and confirmed by duplex ultrasound; 3) death due to procedural complications; 4) limb loss; 5) need for in-hospital salvage revascularization (angioplasty or surgery) of the same treated vessel; or 6) embolic stroke.
The following clinical variables were collected: age, gender, history of diabetes, myocardial infarction (MI), angina, hypertension, hyperlipidemia, smoking (never, prior to the past year and within the past year), prior cerebrovascular events, body mass index, blood pressure at onset of procedure, the presence of peripheral vascular disease with ulceration, recent onset of claudication (Statistical analysis. Descriptive statistics on all variables are initially summarized as proportions or means ± standard deviation. The primary analyses of the study were: 1) to determine the rate of SAEs; and 2) to model the predictors associated with complications, focusing on the revascularization rate and the overall rates.
Kruskal-Wallis tests were used to compare age, body mass indices, dosages and procedure time across treatment cohorts. Kaplan-Meier plots were used to contrast procedure times. ACT categories (highest ACT values were converted to a polynomial variable: 400 seconds) were compared using the exact Jonckheer-Terpstra test of trend across heparin dosage quartiles. Fisher’s exact tests were calculated for unordered categorical variables. Cigarette smoking status was initially described as never, previous and within the past year, but focused on smoking within the past year for modeling. Isotonic regression was used to establish level sets for adverse events as heparin dosages increased. Exact stratified logistic regression analyses were performed to model the need for revascularization salvage procedure and the overall complication rates by potential associated factors while stratifying by physician and simultaneously screening main effects and interactions.
Results
Demographic and health history characteristics of the population studied are shown in Table 1. Seventy-five patients (57.3%) were male. The mean age was 66.4 ± 12.1 years; however, the age distribution is a mixture of distributions (Figure 1). The younger patients had a higher prevalence of almost every cardiovascular historical event. Most notable were the 36.7% with a history of MI who had a median age of 62 years, versus those without an MI who had a median age of 72 years (p = 0.005), and those with hypercholesterolemia who had a median age of 65 years versus those without who had a median age of 73 years (p = 0.005).
There were 73 patients (55.7%) with a documented history of smoking [50 patients (38.2%) smoked within the past year], and 38 (29.0%) were diabetic (57.9% requiring insulin). There were 15 patients (11.5%) with lower leg ulcerations, forty-five patients (34.5%) with a recent onset of claudication ( 400 seconds, respectively. Figure 2 shows that the greater the heparin dosage by quartile, the greater the ACT (p = 0.0016). There is a missing ACT bias, in that ACT is more likely to be measured in the fourth heparin quartile 96.9% of the time as compared to the lower quartiles (83.8%). However, this cannot be attributed to a heparin effect on ACT, as much as to the length of the procedure and the lower ACTs leading to higher dosages of heparin.
The heparin dosing tended to ignore weight, but this created substantial associations between dosage/kg and other factors, including procedure time, diabetes, smoking and ACT. The diabetics were significantly heavier than the nondiabetics (p 1 reported total and major complication rates of 10.5% and 5.0%, respectively, in 295 consecutive patients undergoing lower-limb PTA. In their prospective study, females and patients with total occlusions had higher complication rates than males or patients with stenoses. Morse et al.4 showed an 8.8% significant complication rate in 370 patients treated with lower limb angioplasty. Elderly patients were at significantly increased risk for complications. In 202 PTA procedures, Hasson et al.5 showed a total complication rate of 32.7%. In their series, 1.4% of the complications were amputations and 2.2% were deaths. In multivariate analysis, the most important predictor of the occurrence of a complication, amputation or procedure-related death was the premorbid clinical status of the limb (claudication versus limb threat). Greenfield7 reported a major complication rate of 11.4% (8 out of 70 patients) in the treatment of femoral, popliteal and tibial arteries. In our series, history of current smoking, recent onset of claudication and procedures below the knee were associated with significant adverse clinical events.
Amputation occurred in 1.5% of our patients. In the literature, the amputation rate ranges from 0.6% to approximately 3.0%. In a retrospective study of 71 consecutive patients undergoing infrapopliteal PTAs for limb salvage, Boyer et al.6 reported a global morbidity rate of 16%, including 2 amputations (2.8%), five major vascular complications (7.0%) and 1 death (1.4%). In a prospective audit of 988 peripheral angioplasties, Axisa et al.3 described a 0.6% amputation rate. The amputation rate following angioplasty for critical limb ischemia was 2.2%. In this study, the overall risk of death and/or major complications was 3.5%. Gutteridge et al.2 reported in a non-randomized observational study of 212 patients that amputations were more common in patients undergoing below the inguinal ligament procedures and were more likely to occur in diabetics compared to non-diabetics. In our series, diabetics had fewer complications than non-diabetics. Although this could reflect a selection bias, the diabetics also received significantly less heparin (U/kg) than non-diabetics. In contrast, smokers received more heparin (U/kg) than non-smokers and had a higher complication rate.
In our study, renal insufficiency occurred in 8.3% of patients and 3.1% required dialysis. Since peripheral vascular patients tend to have significant comorbidities for renal failure (renovascular disease, diabetic, hypertensive and congestive heart failure), radiocontrast nephropathy appears to be substantial in those patients. Heparin has been shown to increase bleeding in patients with renal failure; this could have been a contributing factor in increasing bleeding rate.
The optimal ACT or heparin dosing during peripheral vascular interventions has not been well defined. In the coronary literature, and prior to the use of glycoprotein IIb/IIIa inhibitors, Narins and colleagues8 showed a significant inverse relationship between the degree of anticoagulation during angioplasty and the risk of abrupt vessel closure. Ferguson and colleagues9 also showed that a diminished ACT response to an initial bolus of heparin was associated with major in-hospital complications after coronary angioplasty. An “optimal” ACT during coronary interventions, using heparin alone, is generally considered to be between 300–400 seconds. Twenty-nine percent of our patients had a suboptimal ACT ( 400 seconds) than desired, illustrating the difficulty in dosing heparin for the presumed desired ACT range of 300–400 seconds.10,11 In our study, neither the ACT levels nor the heparin dosing (U/kg) were significantly predictive of patients with higher complication rates, although an increased dose of heparin showed a trend toward higher complication rates.
Study limitations. In this retrospective study, we attempted to identify complications from chart review, which has obvious inherent limitations. The sequencing and timing of events associated with SAEs was typically unavailable. Many of the potential explanatory factors are time-dependent instead of proper factors which are known prior to the procedure, such as age, sex, body mass, medical history and which interventionalist is performing the procedure. Time-dependent predictors include procedure time, which can be extended with an SAE, but an extended procedure can result in an SAE; other time-dependent factors include additional dosing of heparin, changes in ACT, occurrence of thrombus, and whether the procedure goes below the knee.
It is possible that not all complications were reported. Also, ACT values were not available on all patients; when this information was available, it was unclear whether ACTs were systematically checked after heparin administration. In addition, this is a single-center experience and might not be shared by other operators. Nevertheless, this report confirms a continued significant morbidity with PTAs despite contemporary techniques. The literature also indicate that the search to replace heparin as a base anticoagulant is timely and needed.
Acknowledgment. The authors wish to thank Peter Teuber, PhD, for his valuable input to this study and the manuscript.
1. Matsi PJ, Manninen HI. Complications of lower-limb percutaneous transluminal angioplasty: A prospective analysis of 410 procedures on 295 consecutive patients. Cardiovasc Intervent Radiol 1998;21:361–366.
2. Gutteridge W, Torrie EP, Galland RB. Cumulative risk of bypass, amputation or death following percutaneous transluminal angioplasty. Eur J Vasc Endovasc Surg 1997;14:134–139.
3. Axisa B, Fishwick G, Bolia A, et al. Complications following peripheral angioplasty. Ann R Coll Surg Engl 2002;84:39–42.
4. Morse MH, Jeans WD, Cole SE, et al. Complications in percutaneous transluminal angioplasty: Relationships with patient age. Br J Radiol 1991;64:5–9.
5. Hasson JE, Acher CW, Wojtowycz M, et al. Lower extremity percutaneous transluminal angioplasty: Multifactorial analysis of morbidity and mortality. Surgery 1990;108:748–752.
6. Boyer L, Therre T, Garcier JM, et al. Infrapopliteal percutaneous transluminal angioplasty for limb salvage. Acta Radiol 2000;41:73–77.
7. Greenfield AJ. Femoral, popliteal and tibial arteries: Percutaneous transluminal angioplasty. Am J Roentgenol 1980;135:927–935.
8. Narins CR, Hillegass WB Jr., Nelson CL, et al. Relation between activated clotting time during angioplasty and abrupt closure. Circulation 1996;93:667–671.
9. Ferguson JJ, Dougherty KG, Gaos CM, et al. Relation between procedural activated coagulation time and outcome after percutaneous transluminal transluminal angioplasty. J Am Coll Cardiol 1994;23:1061–1065.
10. Ogilby JD, Kopelman HA, Klein LW, Agarwal JB. Adequate heparinization during PTCA: Assessment using activated clotting times. Cathet Cardiovasc Diagn 1989;18:206–209.
11. Kleiman NS, Weitz JI. Putting heparin into perspective: Its history and the evolution of its use during percutaneous coronary interventions. J Invas Cardiol 2000;12(Suppl F):20F–26F.
