- Email: [email protected]
Phenylephrine infusion impact on surgical site infections after lower extremity bypass surgery
Phenylephrine infusion impact on surgical site infections after lower extremity bypass surgery Craig Curry, MD,a,b Jens Eldrup-Jorgensen, MD,c Janelle Richard, BA,a Michele C. Siciliano, MD,d and Wendy Y. Craig, PhD,e Portland and Scarborough, Me
ABSTRACT Objective: Lower extremity bypass (LEB) operations have high rates of surgical site infections (SSI). Phenylephrine is a commonly used vasoconstrictor which may reduce skin blood flow and increase the likelihood of SSI in these patients. We studied the potential effect of phenylephrine infusion during LEB surgery on SSI. Methods: LEB cases and their demographic data were identified through the Vascular Quality Initiative registry. SSI in this population was identified using the hospital epidemiology surveillance database. Phenylephrine use in this population was identified through chart review. Results: We identified 699 patients who underwent LEB; 82 (11.7%) developed an SSI, and 244 of 698 (35.0%) were treated with phenylephrine infusion. In bivariate analysis, higher body mass index (28.8 kg/m2 vs 27.3 kg/m2; P ¼ .034), diabetes (14.6% vs 9.4%; P ¼ .035), hypertension (12.6% vs 4.7%; P ¼ .038), groin incision (13.2 vs 5.4%; P ¼ .013) and longer procedure times (17.1% for >220 minutes and 8.9% for #220 minutes; P ¼ .003) were associated with higher rates of SSI. Whereas phenylephrine infusion exhibited a trend toward a higher rate (14.8% vs 9.9%; P ¼ .057). In the logistic regression model, diabetes (odds ratio [OR], 1.8; 95% confidence interval [CI], 1.0-3.2; P ¼ .032), total procedure time (OR, 1.85; 95% CI, 1.1-3.1; P ¼ .026) and vertical groin incision (OR, 2.6; 95% CI, 1.1-6.5; P ¼ .035) were independent predictors of increased SSI rates, whereas body mass index (OR, 1.04; 95% CI, 0.99-1.08; P ¼ .09), hypertension (OR, 2.5; 95% CI, 0.6-10.9; P ¼ .22), and phenylephrine infusion (OR, 1.08; 95% CI, 0.63-1.85; P ¼ .78) were not independent predictors of increased SSI rates. Conclusions: Phenylephrine infusion did not increase the risk of SSI in patients who underwent LEB. (J Vasc Surg 2017;-:1-7.)
Surgical site infections (SSI) result in significant morbidity for patients with additional expenses to the health care system. Vascular surgery has a relatively high rate of SSI and patients undergoing lower extremity bypass (LEB) surgery are at particularly high risk for SSI.1-3 The incidence of SSI after LEB reported in the literature varies from 11% to 31%.4-6 Patient-related variables such as obesity, female gender, and diabetes, and operative variables such as incision site, operative time, blood transfusion, skin preparation, and prophylactic antibiotic use have all been identified as risk factors for SSI in LEB.3-5,7,8 Intraoperative hypotension is common during anesthesia and is frequently treated with vasopressors such as phenylephrine. Fluid management of the vascular patient must be judicious in view of frequent comorbidities, including heart disease and renal insufficiency. In an effort to maintain an acceptable blood pressure without excessive intravenous fluid administration, From the Department of Anesthesiology, Maine Medical Center,a the Spectrum Medical Group,b the Department of Vascular Surgery,c and Department of Surgery,d Maine Medical Center, Portland; and the Maine Medical Center Research Institute, Scarborough.e Author conflict of interest: none. Correspondence: Craig Curry, MD, Department of Anesthesiology, Maine Medical Center, 22 Bramhall St, Portland, ME 04102 (e-mail: curryc@spectrummg. com). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.05.130
phenylephrine is a common supportive agent. However, the use of an alpha constrictor raises concerns about tissue perfusion, which could contribute to infectious complications. In the present study, we evaluate the effect of intraoperative phenylephrine treatment on the incidence of SSI after LEB.
METHODS All patients who underwent LEB (femoral popliteal, femoral tibial, or popliteal tibial bypass for occlusive disease) at Maine Medical Center from January 2010 to September 2015 were eligible for inclusion in the study. Patients were identified retrospectively in the Vascular Quality Initiative (VQI) database, which also provided associated demographic and clinical data, including duration of surgery and number of units transfused.9 For patients who underwent multiple surgical procedures during the study period, we included data from the first encounter only. SSI cases were identified by the hospital epidemiology department using Centers for Disease Control National and Prevention Healthcare Safety Network criteria. This is a surveillance process that includes (a) systematic chart reviews performed by a nurse epidemiologist on all cases of positive microbiology cultures (wound or blood), (b) chart review by an independent nurse data coordinator, (c) hospital readmissions and reoperations, and (d) surgeon self-reporting. The standard survey covered the 30-day period after all operative procedures. All cases that met the National Healthcare Safety Network of the 1
2
Journal of Vascular Surgery
Curry et al
---
Centers for Disease Control National and Prevention definitions were entered into the epidemiology database along with type of surgery, date of surgery, and patient identification (name, date of birth, and hospital record number). These were cross-referenced with the Vascular Quality Initiative cases. Anesthetic records for all LEB cases identified from the Vascular Quality Initiative database were manually reviewed and phenylephrine infusion, fluids, and estimated blood loss (EBL) was recorded. The study was given expedited approval by the Maine Medical Center Institutional Review Board and received a waiver of informed consent according to 45 CFR 46.116(d) and a waiver of authorization according to the Common Rule and HIPAA Privacy Rule 164.512(i)(2). Descriptive data were summarized as mean (95% confidence interval [CI]) or median (range), as appropriate, and categorical data were summarized as frequency (n, %). Continuous data were compared between SSI subgroups by t-test or by Mann-Whitney U test and categorical data were compared by c2 or Fishers Exact test, as appropriate. Logistic regression analysis was used to evaluate potential predictors of infection; SSI was the dependent variable and variables that differed (P < .1) between SSI subgroups in bivariate analysis were entered into the model as independent variables. All analyses were performed using IBM SPSS (Armonk, NY). The study had 80% power (a ¼ .05, twosided test) to detect a 7.5% absolute difference in SSI rates between subjects receiving or not receiving a phenylephrine infusion, assuming a 10% SSI rate in the no-infusion group.
RESULTS Table I shows demographic and clinical characteristics of the study group, stratified by development of an SSI. Age (mean, 68.0 years) and sex (471/699; 67.4% male) were typical for LEB and the racial composition of the study group reflected that of the Maine population (683/691; 97.7% Caucasian). Diabetes (336/688; 48.8%), hypertension (624/688; 90.7%) and current or former smoking (600/687; 87.3%) were frequent and overall, 82 patients (11.7%) developed an SSI. Both diabetes (14.6% vs 9.4%; P ¼ .035) and hypertension (12.6% vs 4.7%; P ¼ .038) were significantly associated with development of an SSI. Those developing an SSI also had a significantly higher body mass index (BMI; 28.8 kg/m2 vs 27.3 kg/m2; P ¼ .034) and, when BMI was stratified to represent obesity status, SSI rates were 53 of 484 (11.0%) at a BMI of #30 kg/m2, 13 of 123 (10.6%) at a BMI of >30-35 kg/m2, 8 of 56 (14.3%) at a BMI of >35-40 kg/m2, and 8 of 23 (34.8%) at a BMI of >40 kg/m2 (P ¼ .006). SSI rates were not significantly higher among patients on dialysis, among those with serum creatinine of >1.8 mg/dL, or among those with prior bypass. Furthermore, there was no difference in the distribution of
2017
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective review of prospectively collected Vascular Quality Initiative data Take Home Message: In 699 patients who underwent lower extremity bypass for critical limb ischemia, 11.7% developed surgical site infection (SSI). Diabetes, total procedure time, and vertical groin incisions were independent predictors of SSI but phenylephrine infusion given for treatment of intraoperative hypotension was not. Recommendation: The authors suggest using phenylephrine if needed to treat intraoperative hypotension of patients who undergo lower extremity bypass for critical limb ischemia, without a risk of increasing SSI.
different indications for surgery between the groups with and without SSI. Prior coronary artery disease or congestive heart failure were both associated with a significantly higher frequency of phenylephrine use (40.7% vs 30.5% [P ¼ .007] and 48% vs 32% [P ¼ .001], respectively). Prior use of beta blockers (36.4% vs 29.4%; P ¼ .14) or angiotensin receptor blockers (41.5% vs 42%; P > .99) had no significant association with the intraoperative use of phenylephrine. There was no increase in SSI frequency among those with prior coronary artery disease (39/300 [13.0%] vs 43/388 [11.1%]; P ¼ .52) or prior congestive heart failure (20/127 [15.7%] vs 62/561 [11.1%]; P ¼ .19). Table II shows procedural data, stratified by development of an SSI. Overall, 244 of 698 patients (35.0%) were treated with phenylephrine infusion; procedure time was 202 minutes (95% CI, 196-208 minutes), 568 of 697 subjects (81.5%) required a groin incision, and 84 of 687 (12.2%) received a blood transfusion during the procedure. In our study group, 107 (15.3%) received no vasopressor, 244 (34.9%) received a phenylephrine infusion, and 573 (82.0%) received a vasopressor bolus. The agents used in bolus dosing were phenylephrine (n ¼ 462), ephedrine (n ¼ 367), vasopressin (n ¼ 18), and epinephrine (n ¼ 4); overall, 347 patients (49.6%) received vasopressor as a bolus only. The SSI rate for those receiving a vasopressor by bolus only was not significantly different compared with those receiving no vasopressor (36/347 [10.4%] vs 9/107 [8.4%]; P ¼ .68); these subgroups were, therefore, combined for comparison with patients receiving a phenylephrine infusion. The SSI rate was slightly, but not significantly, higher among those receiving a phenylephrine infusion, compared with those who did not (14.8% vs 9.9%; P ¼ .057). This represents a 50% relative increase in frequency; in post hoc power analysis, n ¼ 1933 subjects (n ¼ 502 with phenylephrine infusion) would be required to demonstrate statistical significance (80% power, a ¼ .05). SSI
Journal of Vascular Surgery Volume
-,
Number
Curry et al
3
-
Table I. Demographic and clinical characteristics of the study group, overall and stratified by development of a surgical site infection (SSI) SSI a
Variable
All subjects
No. (%)
699
No 617 (88.3)
Yes
P
82 (11.7)
Age, years
68.0 (67.1-68.9)
68.1 (67.2-69.0)
67.2 (64.6-69.9)
.54
BMI, kg/m2b
27.5 (27.0-27.9)
27.3 (26.8-27.8)
28.8 (27.2-30.3)
.034
Sex Male
471
417 (88.5)
54 (11.5)
Female
228
200 (87.7)
28 (12.3)
Caucasianc
683
603 (88.3)
80 (11.7)
87
78 (89.6)
Ex-smoker
324
288 (88.9)
36 (11.1)
Smoker
276
239 (86.6)
37 (13.4)
No
64
61 (95.3)
Yes
624
545 (87.3)
.75 .25
Smokingd No
9 (10.4)
.61
Hypertensione 3 (4.7)
.038
79 (12.6)
Diabetesf No
352
319 (90.6)
33 (9.4)
Yes
336
287 (85.4)
49 (14.6)
No
648
574 (88.6)
74 (11.4)
Yes
34
27 (79.4)
No
601
530 (88.2)
71 (11.8)
Yes
35
32 (91.4)
3 (8.6)
No
502
439 (87.5)
63 (12.5)
Yes
186
167 (89.8)
19 (10.2)
9 (1.6)
1 (1.4)
.035
On dialysisg .18
7 (20.6)
Serum creatinine > 1.8 mg/dLh .76
Prior bypass .48
Indication for surgeryi Asymptomatic PVD Claudication
147 (26.0)
Rest pain
109 (19.3)
13 (18.8)
Tissue loss
242 (42.8)
33 (47.8)
59 (10.4)
5 (7.2)
Acute ischemia
.90j
(24.6)
BMI, Body mass index; DM, diabetes mellitus; NS, not significant; PVD, peripheral vascular disease. a Continuous data are shown as mean (96% confidence interval [CI]) and are compared between SSI subgroups using t-tests; categorical data are shown as number (%) and are compared between SSI subgroups using c2 tests or Fishers exact test, as appropriate. b BMI data were available for n ¼ 686 (n ¼ 604 with no SSI, n ¼ 82 with SSI). c Race data were available for n ¼ 691 (n ¼ 609 with no SSI, n ¼ 82 with SSI). SSI subgroups were compared after grouping subjects as Caucasian or other. The Caucasian subgroup included 1 individual with Hispanic ethnicity; the remaining subjects were black (n ¼ 6), Native American (n ¼ 1), 2 Hawaiian/Pacific Islander (n ¼ 2), and unknown (n ¼ 2). d Data on smoking status were available for n ¼ 687 (n ¼ 605 with no SSI, n ¼ 82 with SSI). e Hypertension and diabetes data were available for n ¼ 688, n ¼ 606 without SSI and n ¼ 82 with SSI. Hypertension was defined as blood pressure $140/90 mmHg or history of hypertension. f Data for diabetic status were available for n ¼ 688 subjects. Among the 336 with diabetes, 33 were treated by diet, 123 with oral medication, and 180 with insulin. g Data available for n ¼ 688; among these, 6 patients with a functioning renal allograft were omitted from the analysis. h Data available for n ¼ 636. i Data were available for n ¼ 635 patients, 69 with SSI and 566 without SSI. j 2 c.
development was, however, significantly associated with both groin incision and longer procedure times. SSI rates were 5.4% among those who did not have a groin incision and 13.2% among those who did (P ¼ .013); among
patients who had groin incisions, the rate of SSI for those with vertical incisions was 59 of 416 (14.2%) compared with 12 of 119 (10.1%) for those with horizontal incisions, respectively (P ¼ .24). Similarly, the SSI rate was 17.1% if
Journal of Vascular Surgery
Curry et al
4
---
2017
Table II. Procedural characteristics of the study group, overall and stratified by development of a surgical site infection (SSI) SSI infection Variablea
All subjects
No.
No
Yes
699
617 (88.3)
82 (11.7)
No
454
409 (90.1)
45 (9.9)
Yes
244
208 (85.2)
36 (14.8)
#220
395
360 (91.1)
35 (8.9)
>220
199
165 (82.9)
34 (17.1)
P
Phenylephrine infusionb .057
Procedure time, minutesc
EBL, mLd Total crystalloids, mLe
213 (193-232)
212 (191-234)
2142 (2057-2226)
216 (172-259)
2126 (2035-2217)
2259 (2020-2498)
.003 .31 .32
Groin incision None
129
122 (94.6)
7 (5.4)
Any
568
493 (86.8)
75 (13.2)
#2
655 (95.3)
576 (87.9)
79 (12.1)
>2
32 (4.7)
30 (93.8)
2 (6.2)
.013
Transfusion, U .57
EBL, Estimated blood loss; NS, not significant. a Continuous data are shown as mean (95% confidence interval [CI]) and are compared between SSI subgroups using t-tests or the Mann-Whitney U test, as appropriate; categorical data are shown as number (%) and are compared between SSI subgroups using c2 tests or Fishers exact test, as appropriate. b Phenylephrine infusion data were available for n ¼ 698; n ¼ 617 without SSI and n ¼ 81 with SSI. c Procedure time was available for n ¼ 594; n ¼ 525 without SSI and n ¼ 69 with SSI. d EBL data were available for n ¼ 695; n ¼ 613 without SSI and n ¼ 82 with SSI. e Total crystalloid data were available for n ¼ 695; n ¼ 614 without SSI and n ¼ 81 with SSI.
Table III. Crystalloid volume, estimated blood loss (EBL), and procedure time, stratified by phenylephrine infusion Phenylephrine infusion Variable
No infusion, No.
Mean (95% CI)
No.
Mean (95% CI)
Pa <.001
EBL, mL
452
191 (166-215)
242
252 (220-284)
Crystalloids, mL
451
2089 (1989-2188)
244
2240 (2085-2394)
Procedure time, minutes
383
191 (184-198)
210
222 (211-232)
.19 <.001
CI, Confidence interval. a Mann-Whitney U test.
procedure time was >220 minutes, compared with 8.9% when procedure time was #220 minutes (P ¼ .003). Table III shows the relationship between intraoperative factors and phenylephrine infusion; both procedure time (222 minutes vs 191 minutes; P < .001) and EBL (252 mL vs 191 mL; P < .001) were significantly higher among those given phenylephrine infusions. To evaluate potential predictors of SSI, we entered BMI, total procedure time (cutoff, 220 minutes), diabetes (yes/ no), hypertension (yes/no), groin incision (by type), and phenylephrine infusion (yes/no) into a logistic regression model, with SSI as the dependent variable. Diabetes (odds ratio [OR], 1.8; 95% CI, 1.0-3.2; P ¼ .032), total procedure time (OR, 1.85; 95% CI, 1.1-3.1; P ¼ .026) and vertical groin incision (OR, 2.6; 95% CI, 1.1-6.5; P ¼ .035) were significant independent predictors of SSI; BMI (OR, 1.04;
95% CI, 0.99-1.08; P ¼ .09), hypertension (OR, 2.5; 95% CI, 0.6-10.9; P ¼ .22), and phenylephrine infusion (OR, 1.08; 95% CI, 0.63-1.85; P ¼ .78) did not contribute significantly to the model. When the logistic regression analysis was repeated after stratifying the phenylephrine variable as no vasopressor, vasopressor bolus only and phenylephrine infusion with or without bolus, compared with the no vasopressor group, the OR was 1.15 (95% CI, 0.48-2.8; P ¼ .75) for vasopressor bolus only and 1.21 (95% CI, 0.49-3.00; P ¼ .68) for phenylephrine infusion. When the analysis was limited to individuals with a groin incision, diabetes (OR, 1.8; 95% CI, 1.0-3.2; P ¼ .048) and total procedure time (OR, 2.0; 1.1-3.4; P ¼ .019) remained significant predictors of SSI. Because there is a relationship between procedure time and phenylephrine infusion in bivariate analysis (221 6 78 minutes with phenylephrine
Journal of Vascular Surgery Volume
-,
Number
Curry et al
5
-
infusion vs 191 6 73 minutes without; P < .001), we repeated the logistic regression analysis omitting procedure time. The OR for infection associated with phenylephrine infusion remained nonsignificant (1.50; 95% CI, 0.93-2.42; P ¼ .095).
DISCUSSION In our patient cohort, we had an incidence of 11% SSI in our study population, which is consistent with those reported in other studies of LEB at 30 days of followup.4,5,10 The study did demonstrate the effects of other variables previously reported to affect SSI, specifically morbid obesity, diabetes, duration of procedure >220 minutes, and groin incision suggesting that this study was adequately powered to detect clinically significant effects. Duration of procedure and vertical groin incision, as well as phenylephrine infusion, are potential surrogate markers for a more complex operation. The rate of SSI observed in the present study was not affected by the use of phenylephrine infusion and, although there was a trend toward a higher SSI rate observed in bivariate analysis, this became less significant after taking covariates into account. As our post hoc power analysis demonstrated, a much larger study might demonstrate statistical significance for this effect in bivariate analysis; however, the effect is small and its further reduction in multivariate analysis suggests that it would likely have limited clinical significance. Phenylephrine increases perfusion pressure through increased arterial and venous tone with little effect on heart rate or myocardial contractility.11 Patients with peripheral vascular disease owing to arteriosclerosis have an increased risk of hypotension with anesthesia and one study reported that 50% are treated with vasopressor during LEB surgery.6,12 Although phenylephrine is effective in increasing mean arterial pressure, it also reduces cutaneous blood flow.13 Decreased cutaneous perfusion is a theoretical contributor to SSI.14 At our own institution, we have developed a preoperative bundle in an attempt to reduce SSI. As we scrutinized the processes of care, we considered the implications of intraoperative hemodynamic status and the use of phenylephrine. The potential for reduced tissue perfusion with vasopressor use has also been theorized to contribute to flap failure in free tissue transfers and anastomosis breakdown on esophagostomies. Reduced flap perfusion with phenylephrine has been shown in animal models.15 However, recent systematic reviews of clinical studies on vasopressor impact on free flap survival by Motakef et al16 and Swanson et al17 found no studies in which vasopressors were found to have a negative impact. Similarly, Pathak et al18 have shown that phenylephrine infusion actually increases blood flow to the stomach during esophagectomy when used to treat hypotension caused by thoracic epidural analgesia.
These studies would corroborate the lack of end-organ effect when phenylephrine is used to reverse hypotension during surgery. There may be systemic, local, and incisional mechanisms that mitigate the vasoconstrictive effect of vasopressors during LEB. Increasing the systemic perfusion pressure may increase skin perfusion when compared with a lower systemic perfusion pressure in untreated hypotension despite cutaneous vasoconstriction. This has been shown in a study of blood flow in a free flap model.19 Phenylephrine has variable effects on skin blood flow in different areas of the body. One study has shown reduced blood flow to the skin on the hand during phenylephrine infusion, while flow to more proximal skin on the forearm was increased, likely owing to a vagal-mediated reflex mechanism.20 If similar differential effects exist in the lower extremity, phenylephrine may actually increase skin perfusion in the groin during LEB. Finally, mechanisms activated by surgical trauma may mitigate reduced skin blood flow with phenylephrine. Raza et al21 have shown increased cutaneous blood flow to the groin incision after LEB. This increase is thought to be related to inflammation from surgical trauma and such a mechanism may be resistant to the vasoconstrictive effects of phenylephrine. Phenylephrine infusion was used frequently in this study population. Its use was associated with longer case times and higher EBL, with a trend toward more crystalloid use. This finding suggests that phenylephrine infusions were likely in part treating hypovolemia, although the actual difference in EBL was small and would have been partially made up by the slightly larger crystalloid dose. It is also likely that phenylephrine infusions were treating hypotension related to general anesthesia, which in part results from reduced circulating epinephrine and norepinephrine.22,23 Such anestheticinduced hypotension is more likely in vascular patients, which may explain the high rate of use in the present study.12 Using vasopressors in this setting may limit excessive crystalloid input, which has been shown to have positive effects in many studies of goal-directed fluid therapy. As a single site study in a racially homogenous population, our study may have limited generalizability. Furthermore, because this was a retrospective study there were some missing medical record data and there was not a standardized protocol for the use of phenylephrine. Administration of phenylephrine was administered based on individual clinical judgement and we do not have data that address the triggers used to initiate an infusion nor the specific physiologic goal of treatment, such as a minimum systolic blood pressure; each of these factors might significantly impact the effect of phenylephrine use on SSI. The dose of phenylephrine administered during infusions was not available and, although it is assumed to be
6
Curry et al
Journal of Vascular Surgery ---
higher than that received in a vasopressor bolus, the actual difference cannot be quantified. Similarly, we could not investigate the potential impact of phenylephrine dose among those receiving an infusion, which may have led us to underestimate the effect of phenylephrine on SSI, because it has differential effects on tissue and organ perfusion at different doses.11 Last, the study was underpowered to detect the observed effect size of phenylephrine on SSI rate (4.9% absolute increase) in bivariate analysis. Larger studies that include phenylephrine dose data may be useful to further define the impact of phenylephrine on SSI in LEB and determine whether it has independent clinical relevance. In summary, the present study found the independent variables associated with SSI are morbid obesity, diabetes, duration of operations of >220 minutes, and groin incision. Phenylephrine given by infusion used for the treatment of hypotension during LEB did not have a significant independent effect on SSI. This project received support in the form of nonclinical office hours for the corresponding author, Dr Curry, from Spectrum Medical Group. Spectrum Medical Group had no involvement in the study design; collection, analysis and interpretation of data; manuscript writing; or the decision to submit the manuscript for publication.
AUTHOR CONTRIBUTIONS Conception and design: CC, JJ Analysis and interpretation: CC, JJ, MS, WC Data collection: MS, JR Writing the article: CC, MS, WC Critical revision of the article: CC, JJ, JR, WC Final approval of the article: CC, JJ, MS, JR, WC Statistical analysis: WC Obtained funding: Not applicable Overall responsibility: CC
REFERENCES 1. Vogel TR, Dombrovskiy VY, Carson JL, Haser PB, Lowry SF, Graham AM. Infectious complications after elective vascular surgical procedures. J Vasc Surg 2010;51:122-9; discussion: 129-30. 2. Turtiainen J, Saimanen E, Partio T, Karkkainen J, Kiviniemi V, Makinen K, et al. Surgical wound infections after vascular surgery: prospective multicenter observational study. Scand J Surg 2010;99:167-72. 3. Bandyk DF. Vascular surgical site infection: risk factors and preventive measures. Semin Vasc Surg 2008;21:119-23. 4. Giles KA, Hamdan AD, Pomposelli FB, Wyers MC, Siracuse JJ, Schermerhorn ML. Body mass index: surgical site infections and mortality after lower extremity bypass from the National Surgical Quality Improvement Program 2005-2007. Ann Vasc Surg 2010;24:48-56. 5. Greenblatt DY, Rajamanickam V, Mell MW. Predictors of surgical site infection after open lower extremity revascularization. J Vasc Surg 2011;54:433-9.
2017
6. Kuy S, Dua A, Desai S, Dua A, Patel B, Tondravi N, et al. Surgical site infections after lower extremity revascularization procedures involving groin incisions. Ann Vasc Surg 2014;28: 53-8. 7. Kalish JA, Farber A, Homa K, Trinidad M, Beck A, Davies MG, et al. Factors associated with surgical site infection after lower extremity bypass in the Society for Vascular Surgery (SVS) Vascular Quality Initiative (VQI). J Vasc Surg 2014;60: 1238-46. 8. Teixeira PG, Woo K, Weaver FA, Rowe VL. Vein harvesting technique for infrainguinal arterial bypass with great saphenous vein and its association with surgical site infection and graft patency. J Vasc Surg 2015;61:1264-71.e2. 9. Cronenwett JL, Likosky DS, Russell MT, Eldrup-Jorgensen J, Stanley AC, Nolan BW. A regional registry for quality assurance and improvement: the Vascular Study Group of Northern New England (VSGNNE). J Vasc Surg 2007;46: 1093-101; discussion: 1101-2. 10. Tan TW, Kalish JA, Hamburg NM, Rybin D, Doros G, Eberhardt RT, et al. Shorter duration of femoral-popliteal bypass is associated with decreased surgical site infection and shorter hospital length of stay. J Am Coll Surg 2012;215: 512-8. 11. Thiele RH, Nemergut EC, Lynch C 3rd. The physiologic implications of isolated alpha(1) adrenergic stimulation. Anesth Analg 2011;113:284-96. 12. Morimoto Y, Yamagata K, Hanamoto H, Boku A, Kudo C, Yokoe C, et al. Arteriosclerosis can predict hypotension during anesthesia induction in patients 40 years and older. J Clin Anesthes 2014. [Epub ahead of print]. 13. Yamazaki F, Yuge N. Limb-specific differences in the skin vascular responsiveness to adrenergic agonists. J Appl Physiol 2011;111:170-6. 14. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH 3rd, Jensen JA, et al. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg 1997;132:997-1004; discussion: 1005. 15. Lecoq JP, Joris JL, Nelissen XP, Lamy ML, Heymans OY. Effect of adrenergic stimulation on cutaneous microcirculation immediately after surgical adventitiectomy in a rat skin flap model. Microsurgery 2008;28:480-6. 16. Motakef S, Mountziaris PM, Ismail IK, Agag RL, Patel A. Emerging paradigms in perioperative management for microsurgical free tissue transfer: review of the literature and evidence-based guidelines. Plast Reconstr Surg 2015;135: 290-9. 17. Swanson EW, Cheng HT, Susarla SM, Yalanis GC, Lough DM, Johnson O 3rd, et al. Intraoperative use of vasopressors is safe in head and neck free tissue transfer. J Reconstr Microsurg 2016;32:87-93. 18. Pathak D, Pennefather SH, Russell GN, Al Rawi O, Dave IC, Gilby S, et al. Phenylephrine infusion improves blood flow to the stomach during oesophagectomy in the presence of a thoracic epidural analgesia. Eur J Cardiothorac Surg 2013;44: 130-3. 19. Banic A, Krejci V, Erni D, Petersen-Felix S, Sigurdsson GH. Effects of extradural anesthesia on microcirculatory blood flow in free latissimus dorsi musculocutaneous flaps in pigs. Plast Reconstr Surg 1997;100:945-55; discussion: 956. 20. Silverman DG, Jotkowitz AB, Freemer M, Gutter V, O’Connor TZ, Braverman IM. Peripheral assessment of phenylephrine-induced vasoconstriction by laser Doppler flowmetry and its potential relevance to homeostatic mechanisms. Circulation 1994;90:23-6. 21. Raza Z, Newton DJ, Harrison DK, McCollum PT, Stonebridge PA. Disruption of skin perfusion following
Journal of Vascular Surgery Volume
-,
Number
Curry et al
7
-
longitudinal groin incision for infrainguinal bypass surgery. Eur J Vasc Endovasc Surg 1999;17:5-8. 22. Goldmann A, Hoehne C, Fritz GA, Unger J, Ahlers O, Nachtigall I, et al. Combined vs. isoflurane/fentanyl anesthesia for major abdominal surgery: effects on hormones and hemodynamics. Med Sci Monit 2008;14:CR445-52.
23. Marty J, Gauzit R, Lefevre P, Couderc E, Farinotti R, Henzel C, et al. Effects of diazepam and midazolam on baroreflex control of heart rate and on sympathetic activity in humans. Anesth Analg 1986;65:113-9. Submitted Jan 24, 2017; accepted May 24, 2017.
ABSTRACT Objective: Lower extremity bypass (LEB) operations have high rates of surgical site infections (SSI). Phenylephrine is a commonly used vasoconstrictor which may reduce skin blood flow and increase the likelihood of SSI in these patients. We studied the potential effect of phenylephrine infusion during LEB surgery on SSI. Methods: LEB cases and their demographic data were identified through the Vascular Quality Initiative registry. SSI in this population was identified using the hospital epidemiology surveillance database. Phenylephrine use in this population was identified through chart review. Results: We identified 699 patients who underwent LEB; 82 (11.7%) developed an SSI, and 244 of 698 (35.0%) were treated with phenylephrine infusion. In bivariate analysis, higher body mass index (28.8 kg/m2 vs 27.3 kg/m2; P ¼ .034), diabetes (14.6% vs 9.4%; P ¼ .035), hypertension (12.6% vs 4.7%; P ¼ .038), groin incision (13.2 vs 5.4%; P ¼ .013) and longer procedure times (17.1% for >220 minutes and 8.9% for #220 minutes; P ¼ .003) were associated with higher rates of SSI. Whereas phenylephrine infusion exhibited a trend toward a higher rate (14.8% vs 9.9%; P ¼ .057). In the logistic regression model, diabetes (odds ratio [OR], 1.8; 95% confidence interval [CI], 1.0-3.2; P ¼ .032), total procedure time (OR, 1.85; 95% CI, 1.1-3.1; P ¼ .026) and vertical groin incision (OR, 2.6; 95% CI, 1.1-6.5; P ¼ .035) were independent predictors of increased SSI rates, whereas body mass index (OR, 1.04; 95% CI, 0.99-1.08; P ¼ .09), hypertension (OR, 2.5; 95% CI, 0.6-10.9; P ¼ .22), and phenylephrine infusion (OR, 1.08; 95% CI, 0.63-1.85; P ¼ .78) were not independent predictors of increased SSI rates. Conclusions: Phenylephrine infusion did not increase the risk of SSI in patients who underwent LEB. (J Vasc Surg 2017;-:1-7.)
Surgical site infections (SSI) result in significant morbidity for patients with additional expenses to the health care system. Vascular surgery has a relatively high rate of SSI and patients undergoing lower extremity bypass (LEB) surgery are at particularly high risk for SSI.1-3 The incidence of SSI after LEB reported in the literature varies from 11% to 31%.4-6 Patient-related variables such as obesity, female gender, and diabetes, and operative variables such as incision site, operative time, blood transfusion, skin preparation, and prophylactic antibiotic use have all been identified as risk factors for SSI in LEB.3-5,7,8 Intraoperative hypotension is common during anesthesia and is frequently treated with vasopressors such as phenylephrine. Fluid management of the vascular patient must be judicious in view of frequent comorbidities, including heart disease and renal insufficiency. In an effort to maintain an acceptable blood pressure without excessive intravenous fluid administration, From the Department of Anesthesiology, Maine Medical Center,a the Spectrum Medical Group,b the Department of Vascular Surgery,c and Department of Surgery,d Maine Medical Center, Portland; and the Maine Medical Center Research Institute, Scarborough.e Author conflict of interest: none. Correspondence: Craig Curry, MD, Department of Anesthesiology, Maine Medical Center, 22 Bramhall St, Portland, ME 04102 (e-mail: curryc@spectrummg. com). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.05.130
phenylephrine is a common supportive agent. However, the use of an alpha constrictor raises concerns about tissue perfusion, which could contribute to infectious complications. In the present study, we evaluate the effect of intraoperative phenylephrine treatment on the incidence of SSI after LEB.
METHODS All patients who underwent LEB (femoral popliteal, femoral tibial, or popliteal tibial bypass for occlusive disease) at Maine Medical Center from January 2010 to September 2015 were eligible for inclusion in the study. Patients were identified retrospectively in the Vascular Quality Initiative (VQI) database, which also provided associated demographic and clinical data, including duration of surgery and number of units transfused.9 For patients who underwent multiple surgical procedures during the study period, we included data from the first encounter only. SSI cases were identified by the hospital epidemiology department using Centers for Disease Control National and Prevention Healthcare Safety Network criteria. This is a surveillance process that includes (a) systematic chart reviews performed by a nurse epidemiologist on all cases of positive microbiology cultures (wound or blood), (b) chart review by an independent nurse data coordinator, (c) hospital readmissions and reoperations, and (d) surgeon self-reporting. The standard survey covered the 30-day period after all operative procedures. All cases that met the National Healthcare Safety Network of the 1
2
Journal of Vascular Surgery
Curry et al
---
Centers for Disease Control National and Prevention definitions were entered into the epidemiology database along with type of surgery, date of surgery, and patient identification (name, date of birth, and hospital record number). These were cross-referenced with the Vascular Quality Initiative cases. Anesthetic records for all LEB cases identified from the Vascular Quality Initiative database were manually reviewed and phenylephrine infusion, fluids, and estimated blood loss (EBL) was recorded. The study was given expedited approval by the Maine Medical Center Institutional Review Board and received a waiver of informed consent according to 45 CFR 46.116(d) and a waiver of authorization according to the Common Rule and HIPAA Privacy Rule 164.512(i)(2). Descriptive data were summarized as mean (95% confidence interval [CI]) or median (range), as appropriate, and categorical data were summarized as frequency (n, %). Continuous data were compared between SSI subgroups by t-test or by Mann-Whitney U test and categorical data were compared by c2 or Fishers Exact test, as appropriate. Logistic regression analysis was used to evaluate potential predictors of infection; SSI was the dependent variable and variables that differed (P < .1) between SSI subgroups in bivariate analysis were entered into the model as independent variables. All analyses were performed using IBM SPSS (Armonk, NY). The study had 80% power (a ¼ .05, twosided test) to detect a 7.5% absolute difference in SSI rates between subjects receiving or not receiving a phenylephrine infusion, assuming a 10% SSI rate in the no-infusion group.
RESULTS Table I shows demographic and clinical characteristics of the study group, stratified by development of an SSI. Age (mean, 68.0 years) and sex (471/699; 67.4% male) were typical for LEB and the racial composition of the study group reflected that of the Maine population (683/691; 97.7% Caucasian). Diabetes (336/688; 48.8%), hypertension (624/688; 90.7%) and current or former smoking (600/687; 87.3%) were frequent and overall, 82 patients (11.7%) developed an SSI. Both diabetes (14.6% vs 9.4%; P ¼ .035) and hypertension (12.6% vs 4.7%; P ¼ .038) were significantly associated with development of an SSI. Those developing an SSI also had a significantly higher body mass index (BMI; 28.8 kg/m2 vs 27.3 kg/m2; P ¼ .034) and, when BMI was stratified to represent obesity status, SSI rates were 53 of 484 (11.0%) at a BMI of #30 kg/m2, 13 of 123 (10.6%) at a BMI of >30-35 kg/m2, 8 of 56 (14.3%) at a BMI of >35-40 kg/m2, and 8 of 23 (34.8%) at a BMI of >40 kg/m2 (P ¼ .006). SSI rates were not significantly higher among patients on dialysis, among those with serum creatinine of >1.8 mg/dL, or among those with prior bypass. Furthermore, there was no difference in the distribution of
2017
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective review of prospectively collected Vascular Quality Initiative data Take Home Message: In 699 patients who underwent lower extremity bypass for critical limb ischemia, 11.7% developed surgical site infection (SSI). Diabetes, total procedure time, and vertical groin incisions were independent predictors of SSI but phenylephrine infusion given for treatment of intraoperative hypotension was not. Recommendation: The authors suggest using phenylephrine if needed to treat intraoperative hypotension of patients who undergo lower extremity bypass for critical limb ischemia, without a risk of increasing SSI.
different indications for surgery between the groups with and without SSI. Prior coronary artery disease or congestive heart failure were both associated with a significantly higher frequency of phenylephrine use (40.7% vs 30.5% [P ¼ .007] and 48% vs 32% [P ¼ .001], respectively). Prior use of beta blockers (36.4% vs 29.4%; P ¼ .14) or angiotensin receptor blockers (41.5% vs 42%; P > .99) had no significant association with the intraoperative use of phenylephrine. There was no increase in SSI frequency among those with prior coronary artery disease (39/300 [13.0%] vs 43/388 [11.1%]; P ¼ .52) or prior congestive heart failure (20/127 [15.7%] vs 62/561 [11.1%]; P ¼ .19). Table II shows procedural data, stratified by development of an SSI. Overall, 244 of 698 patients (35.0%) were treated with phenylephrine infusion; procedure time was 202 minutes (95% CI, 196-208 minutes), 568 of 697 subjects (81.5%) required a groin incision, and 84 of 687 (12.2%) received a blood transfusion during the procedure. In our study group, 107 (15.3%) received no vasopressor, 244 (34.9%) received a phenylephrine infusion, and 573 (82.0%) received a vasopressor bolus. The agents used in bolus dosing were phenylephrine (n ¼ 462), ephedrine (n ¼ 367), vasopressin (n ¼ 18), and epinephrine (n ¼ 4); overall, 347 patients (49.6%) received vasopressor as a bolus only. The SSI rate for those receiving a vasopressor by bolus only was not significantly different compared with those receiving no vasopressor (36/347 [10.4%] vs 9/107 [8.4%]; P ¼ .68); these subgroups were, therefore, combined for comparison with patients receiving a phenylephrine infusion. The SSI rate was slightly, but not significantly, higher among those receiving a phenylephrine infusion, compared with those who did not (14.8% vs 9.9%; P ¼ .057). This represents a 50% relative increase in frequency; in post hoc power analysis, n ¼ 1933 subjects (n ¼ 502 with phenylephrine infusion) would be required to demonstrate statistical significance (80% power, a ¼ .05). SSI
Journal of Vascular Surgery Volume
-,
Number
Curry et al
3
-
Table I. Demographic and clinical characteristics of the study group, overall and stratified by development of a surgical site infection (SSI) SSI a
Variable
All subjects
No. (%)
699
No 617 (88.3)
Yes
P
82 (11.7)
Age, years
68.0 (67.1-68.9)
68.1 (67.2-69.0)
67.2 (64.6-69.9)
.54
BMI, kg/m2b
27.5 (27.0-27.9)
27.3 (26.8-27.8)
28.8 (27.2-30.3)
.034
Sex Male
471
417 (88.5)
54 (11.5)
Female
228
200 (87.7)
28 (12.3)
Caucasianc
683
603 (88.3)
80 (11.7)
87
78 (89.6)
Ex-smoker
324
288 (88.9)
36 (11.1)
Smoker
276
239 (86.6)
37 (13.4)
No
64
61 (95.3)
Yes
624
545 (87.3)
.75 .25
Smokingd No
9 (10.4)
.61
Hypertensione 3 (4.7)
.038
79 (12.6)
Diabetesf No
352
319 (90.6)
33 (9.4)
Yes
336
287 (85.4)
49 (14.6)
No
648
574 (88.6)
74 (11.4)
Yes
34
27 (79.4)
No
601
530 (88.2)
71 (11.8)
Yes
35
32 (91.4)
3 (8.6)
No
502
439 (87.5)
63 (12.5)
Yes
186
167 (89.8)
19 (10.2)
9 (1.6)
1 (1.4)
.035
On dialysisg .18
7 (20.6)
Serum creatinine > 1.8 mg/dLh .76
Prior bypass .48
Indication for surgeryi Asymptomatic PVD Claudication
147 (26.0)
Rest pain
109 (19.3)
13 (18.8)
Tissue loss
242 (42.8)
33 (47.8)
59 (10.4)
5 (7.2)
Acute ischemia
.90j
(24.6)
BMI, Body mass index; DM, diabetes mellitus; NS, not significant; PVD, peripheral vascular disease. a Continuous data are shown as mean (96% confidence interval [CI]) and are compared between SSI subgroups using t-tests; categorical data are shown as number (%) and are compared between SSI subgroups using c2 tests or Fishers exact test, as appropriate. b BMI data were available for n ¼ 686 (n ¼ 604 with no SSI, n ¼ 82 with SSI). c Race data were available for n ¼ 691 (n ¼ 609 with no SSI, n ¼ 82 with SSI). SSI subgroups were compared after grouping subjects as Caucasian or other. The Caucasian subgroup included 1 individual with Hispanic ethnicity; the remaining subjects were black (n ¼ 6), Native American (n ¼ 1), 2 Hawaiian/Pacific Islander (n ¼ 2), and unknown (n ¼ 2). d Data on smoking status were available for n ¼ 687 (n ¼ 605 with no SSI, n ¼ 82 with SSI). e Hypertension and diabetes data were available for n ¼ 688, n ¼ 606 without SSI and n ¼ 82 with SSI. Hypertension was defined as blood pressure $140/90 mmHg or history of hypertension. f Data for diabetic status were available for n ¼ 688 subjects. Among the 336 with diabetes, 33 were treated by diet, 123 with oral medication, and 180 with insulin. g Data available for n ¼ 688; among these, 6 patients with a functioning renal allograft were omitted from the analysis. h Data available for n ¼ 636. i Data were available for n ¼ 635 patients, 69 with SSI and 566 without SSI. j 2 c.
development was, however, significantly associated with both groin incision and longer procedure times. SSI rates were 5.4% among those who did not have a groin incision and 13.2% among those who did (P ¼ .013); among
patients who had groin incisions, the rate of SSI for those with vertical incisions was 59 of 416 (14.2%) compared with 12 of 119 (10.1%) for those with horizontal incisions, respectively (P ¼ .24). Similarly, the SSI rate was 17.1% if
Journal of Vascular Surgery
Curry et al
4
---
2017
Table II. Procedural characteristics of the study group, overall and stratified by development of a surgical site infection (SSI) SSI infection Variablea
All subjects
No.
No
Yes
699
617 (88.3)
82 (11.7)
No
454
409 (90.1)
45 (9.9)
Yes
244
208 (85.2)
36 (14.8)
#220
395
360 (91.1)
35 (8.9)
>220
199
165 (82.9)
34 (17.1)
P
Phenylephrine infusionb .057
Procedure time, minutesc
EBL, mLd Total crystalloids, mLe
213 (193-232)
212 (191-234)
2142 (2057-2226)
216 (172-259)
2126 (2035-2217)
2259 (2020-2498)
.003 .31 .32
Groin incision None
129
122 (94.6)
7 (5.4)
Any
568
493 (86.8)
75 (13.2)
#2
655 (95.3)
576 (87.9)
79 (12.1)
>2
32 (4.7)
30 (93.8)
2 (6.2)
.013
Transfusion, U .57
EBL, Estimated blood loss; NS, not significant. a Continuous data are shown as mean (95% confidence interval [CI]) and are compared between SSI subgroups using t-tests or the Mann-Whitney U test, as appropriate; categorical data are shown as number (%) and are compared between SSI subgroups using c2 tests or Fishers exact test, as appropriate. b Phenylephrine infusion data were available for n ¼ 698; n ¼ 617 without SSI and n ¼ 81 with SSI. c Procedure time was available for n ¼ 594; n ¼ 525 without SSI and n ¼ 69 with SSI. d EBL data were available for n ¼ 695; n ¼ 613 without SSI and n ¼ 82 with SSI. e Total crystalloid data were available for n ¼ 695; n ¼ 614 without SSI and n ¼ 81 with SSI.
Table III. Crystalloid volume, estimated blood loss (EBL), and procedure time, stratified by phenylephrine infusion Phenylephrine infusion Variable
No infusion, No.
Mean (95% CI)
No.
Mean (95% CI)
Pa <.001
EBL, mL
452
191 (166-215)
242
252 (220-284)
Crystalloids, mL
451
2089 (1989-2188)
244
2240 (2085-2394)
Procedure time, minutes
383
191 (184-198)
210
222 (211-232)
.19 <.001
CI, Confidence interval. a Mann-Whitney U test.
procedure time was >220 minutes, compared with 8.9% when procedure time was #220 minutes (P ¼ .003). Table III shows the relationship between intraoperative factors and phenylephrine infusion; both procedure time (222 minutes vs 191 minutes; P < .001) and EBL (252 mL vs 191 mL; P < .001) were significantly higher among those given phenylephrine infusions. To evaluate potential predictors of SSI, we entered BMI, total procedure time (cutoff, 220 minutes), diabetes (yes/ no), hypertension (yes/no), groin incision (by type), and phenylephrine infusion (yes/no) into a logistic regression model, with SSI as the dependent variable. Diabetes (odds ratio [OR], 1.8; 95% CI, 1.0-3.2; P ¼ .032), total procedure time (OR, 1.85; 95% CI, 1.1-3.1; P ¼ .026) and vertical groin incision (OR, 2.6; 95% CI, 1.1-6.5; P ¼ .035) were significant independent predictors of SSI; BMI (OR, 1.04;
95% CI, 0.99-1.08; P ¼ .09), hypertension (OR, 2.5; 95% CI, 0.6-10.9; P ¼ .22), and phenylephrine infusion (OR, 1.08; 95% CI, 0.63-1.85; P ¼ .78) did not contribute significantly to the model. When the logistic regression analysis was repeated after stratifying the phenylephrine variable as no vasopressor, vasopressor bolus only and phenylephrine infusion with or without bolus, compared with the no vasopressor group, the OR was 1.15 (95% CI, 0.48-2.8; P ¼ .75) for vasopressor bolus only and 1.21 (95% CI, 0.49-3.00; P ¼ .68) for phenylephrine infusion. When the analysis was limited to individuals with a groin incision, diabetes (OR, 1.8; 95% CI, 1.0-3.2; P ¼ .048) and total procedure time (OR, 2.0; 1.1-3.4; P ¼ .019) remained significant predictors of SSI. Because there is a relationship between procedure time and phenylephrine infusion in bivariate analysis (221 6 78 minutes with phenylephrine
Journal of Vascular Surgery Volume
-,
Number
Curry et al
5
-
infusion vs 191 6 73 minutes without; P < .001), we repeated the logistic regression analysis omitting procedure time. The OR for infection associated with phenylephrine infusion remained nonsignificant (1.50; 95% CI, 0.93-2.42; P ¼ .095).
DISCUSSION In our patient cohort, we had an incidence of 11% SSI in our study population, which is consistent with those reported in other studies of LEB at 30 days of followup.4,5,10 The study did demonstrate the effects of other variables previously reported to affect SSI, specifically morbid obesity, diabetes, duration of procedure >220 minutes, and groin incision suggesting that this study was adequately powered to detect clinically significant effects. Duration of procedure and vertical groin incision, as well as phenylephrine infusion, are potential surrogate markers for a more complex operation. The rate of SSI observed in the present study was not affected by the use of phenylephrine infusion and, although there was a trend toward a higher SSI rate observed in bivariate analysis, this became less significant after taking covariates into account. As our post hoc power analysis demonstrated, a much larger study might demonstrate statistical significance for this effect in bivariate analysis; however, the effect is small and its further reduction in multivariate analysis suggests that it would likely have limited clinical significance. Phenylephrine increases perfusion pressure through increased arterial and venous tone with little effect on heart rate or myocardial contractility.11 Patients with peripheral vascular disease owing to arteriosclerosis have an increased risk of hypotension with anesthesia and one study reported that 50% are treated with vasopressor during LEB surgery.6,12 Although phenylephrine is effective in increasing mean arterial pressure, it also reduces cutaneous blood flow.13 Decreased cutaneous perfusion is a theoretical contributor to SSI.14 At our own institution, we have developed a preoperative bundle in an attempt to reduce SSI. As we scrutinized the processes of care, we considered the implications of intraoperative hemodynamic status and the use of phenylephrine. The potential for reduced tissue perfusion with vasopressor use has also been theorized to contribute to flap failure in free tissue transfers and anastomosis breakdown on esophagostomies. Reduced flap perfusion with phenylephrine has been shown in animal models.15 However, recent systematic reviews of clinical studies on vasopressor impact on free flap survival by Motakef et al16 and Swanson et al17 found no studies in which vasopressors were found to have a negative impact. Similarly, Pathak et al18 have shown that phenylephrine infusion actually increases blood flow to the stomach during esophagectomy when used to treat hypotension caused by thoracic epidural analgesia.
These studies would corroborate the lack of end-organ effect when phenylephrine is used to reverse hypotension during surgery. There may be systemic, local, and incisional mechanisms that mitigate the vasoconstrictive effect of vasopressors during LEB. Increasing the systemic perfusion pressure may increase skin perfusion when compared with a lower systemic perfusion pressure in untreated hypotension despite cutaneous vasoconstriction. This has been shown in a study of blood flow in a free flap model.19 Phenylephrine has variable effects on skin blood flow in different areas of the body. One study has shown reduced blood flow to the skin on the hand during phenylephrine infusion, while flow to more proximal skin on the forearm was increased, likely owing to a vagal-mediated reflex mechanism.20 If similar differential effects exist in the lower extremity, phenylephrine may actually increase skin perfusion in the groin during LEB. Finally, mechanisms activated by surgical trauma may mitigate reduced skin blood flow with phenylephrine. Raza et al21 have shown increased cutaneous blood flow to the groin incision after LEB. This increase is thought to be related to inflammation from surgical trauma and such a mechanism may be resistant to the vasoconstrictive effects of phenylephrine. Phenylephrine infusion was used frequently in this study population. Its use was associated with longer case times and higher EBL, with a trend toward more crystalloid use. This finding suggests that phenylephrine infusions were likely in part treating hypovolemia, although the actual difference in EBL was small and would have been partially made up by the slightly larger crystalloid dose. It is also likely that phenylephrine infusions were treating hypotension related to general anesthesia, which in part results from reduced circulating epinephrine and norepinephrine.22,23 Such anestheticinduced hypotension is more likely in vascular patients, which may explain the high rate of use in the present study.12 Using vasopressors in this setting may limit excessive crystalloid input, which has been shown to have positive effects in many studies of goal-directed fluid therapy. As a single site study in a racially homogenous population, our study may have limited generalizability. Furthermore, because this was a retrospective study there were some missing medical record data and there was not a standardized protocol for the use of phenylephrine. Administration of phenylephrine was administered based on individual clinical judgement and we do not have data that address the triggers used to initiate an infusion nor the specific physiologic goal of treatment, such as a minimum systolic blood pressure; each of these factors might significantly impact the effect of phenylephrine use on SSI. The dose of phenylephrine administered during infusions was not available and, although it is assumed to be
6
Curry et al
Journal of Vascular Surgery ---
higher than that received in a vasopressor bolus, the actual difference cannot be quantified. Similarly, we could not investigate the potential impact of phenylephrine dose among those receiving an infusion, which may have led us to underestimate the effect of phenylephrine on SSI, because it has differential effects on tissue and organ perfusion at different doses.11 Last, the study was underpowered to detect the observed effect size of phenylephrine on SSI rate (4.9% absolute increase) in bivariate analysis. Larger studies that include phenylephrine dose data may be useful to further define the impact of phenylephrine on SSI in LEB and determine whether it has independent clinical relevance. In summary, the present study found the independent variables associated with SSI are morbid obesity, diabetes, duration of operations of >220 minutes, and groin incision. Phenylephrine given by infusion used for the treatment of hypotension during LEB did not have a significant independent effect on SSI. This project received support in the form of nonclinical office hours for the corresponding author, Dr Curry, from Spectrum Medical Group. Spectrum Medical Group had no involvement in the study design; collection, analysis and interpretation of data; manuscript writing; or the decision to submit the manuscript for publication.
AUTHOR CONTRIBUTIONS Conception and design: CC, JJ Analysis and interpretation: CC, JJ, MS, WC Data collection: MS, JR Writing the article: CC, MS, WC Critical revision of the article: CC, JJ, JR, WC Final approval of the article: CC, JJ, MS, JR, WC Statistical analysis: WC Obtained funding: Not applicable Overall responsibility: CC
REFERENCES 1. Vogel TR, Dombrovskiy VY, Carson JL, Haser PB, Lowry SF, Graham AM. Infectious complications after elective vascular surgical procedures. J Vasc Surg 2010;51:122-9; discussion: 129-30. 2. Turtiainen J, Saimanen E, Partio T, Karkkainen J, Kiviniemi V, Makinen K, et al. Surgical wound infections after vascular surgery: prospective multicenter observational study. Scand J Surg 2010;99:167-72. 3. Bandyk DF. Vascular surgical site infection: risk factors and preventive measures. Semin Vasc Surg 2008;21:119-23. 4. Giles KA, Hamdan AD, Pomposelli FB, Wyers MC, Siracuse JJ, Schermerhorn ML. Body mass index: surgical site infections and mortality after lower extremity bypass from the National Surgical Quality Improvement Program 2005-2007. Ann Vasc Surg 2010;24:48-56. 5. Greenblatt DY, Rajamanickam V, Mell MW. Predictors of surgical site infection after open lower extremity revascularization. J Vasc Surg 2011;54:433-9.
2017
6. Kuy S, Dua A, Desai S, Dua A, Patel B, Tondravi N, et al. Surgical site infections after lower extremity revascularization procedures involving groin incisions. Ann Vasc Surg 2014;28: 53-8. 7. Kalish JA, Farber A, Homa K, Trinidad M, Beck A, Davies MG, et al. Factors associated with surgical site infection after lower extremity bypass in the Society for Vascular Surgery (SVS) Vascular Quality Initiative (VQI). J Vasc Surg 2014;60: 1238-46. 8. Teixeira PG, Woo K, Weaver FA, Rowe VL. Vein harvesting technique for infrainguinal arterial bypass with great saphenous vein and its association with surgical site infection and graft patency. J Vasc Surg 2015;61:1264-71.e2. 9. Cronenwett JL, Likosky DS, Russell MT, Eldrup-Jorgensen J, Stanley AC, Nolan BW. A regional registry for quality assurance and improvement: the Vascular Study Group of Northern New England (VSGNNE). J Vasc Surg 2007;46: 1093-101; discussion: 1101-2. 10. Tan TW, Kalish JA, Hamburg NM, Rybin D, Doros G, Eberhardt RT, et al. Shorter duration of femoral-popliteal bypass is associated with decreased surgical site infection and shorter hospital length of stay. J Am Coll Surg 2012;215: 512-8. 11. Thiele RH, Nemergut EC, Lynch C 3rd. The physiologic implications of isolated alpha(1) adrenergic stimulation. Anesth Analg 2011;113:284-96. 12. Morimoto Y, Yamagata K, Hanamoto H, Boku A, Kudo C, Yokoe C, et al. Arteriosclerosis can predict hypotension during anesthesia induction in patients 40 years and older. J Clin Anesthes 2014. [Epub ahead of print]. 13. Yamazaki F, Yuge N. Limb-specific differences in the skin vascular responsiveness to adrenergic agonists. J Appl Physiol 2011;111:170-6. 14. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH 3rd, Jensen JA, et al. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg 1997;132:997-1004; discussion: 1005. 15. Lecoq JP, Joris JL, Nelissen XP, Lamy ML, Heymans OY. Effect of adrenergic stimulation on cutaneous microcirculation immediately after surgical adventitiectomy in a rat skin flap model. Microsurgery 2008;28:480-6. 16. Motakef S, Mountziaris PM, Ismail IK, Agag RL, Patel A. Emerging paradigms in perioperative management for microsurgical free tissue transfer: review of the literature and evidence-based guidelines. Plast Reconstr Surg 2015;135: 290-9. 17. Swanson EW, Cheng HT, Susarla SM, Yalanis GC, Lough DM, Johnson O 3rd, et al. Intraoperative use of vasopressors is safe in head and neck free tissue transfer. J Reconstr Microsurg 2016;32:87-93. 18. Pathak D, Pennefather SH, Russell GN, Al Rawi O, Dave IC, Gilby S, et al. Phenylephrine infusion improves blood flow to the stomach during oesophagectomy in the presence of a thoracic epidural analgesia. Eur J Cardiothorac Surg 2013;44: 130-3. 19. Banic A, Krejci V, Erni D, Petersen-Felix S, Sigurdsson GH. Effects of extradural anesthesia on microcirculatory blood flow in free latissimus dorsi musculocutaneous flaps in pigs. Plast Reconstr Surg 1997;100:945-55; discussion: 956. 20. Silverman DG, Jotkowitz AB, Freemer M, Gutter V, O’Connor TZ, Braverman IM. Peripheral assessment of phenylephrine-induced vasoconstriction by laser Doppler flowmetry and its potential relevance to homeostatic mechanisms. Circulation 1994;90:23-6. 21. Raza Z, Newton DJ, Harrison DK, McCollum PT, Stonebridge PA. Disruption of skin perfusion following
Journal of Vascular Surgery Volume
-,
Number
Curry et al
7
-
longitudinal groin incision for infrainguinal bypass surgery. Eur J Vasc Endovasc Surg 1999;17:5-8. 22. Goldmann A, Hoehne C, Fritz GA, Unger J, Ahlers O, Nachtigall I, et al. Combined vs. isoflurane/fentanyl anesthesia for major abdominal surgery: effects on hormones and hemodynamics. Med Sci Monit 2008;14:CR445-52.
23. Marty J, Gauzit R, Lefevre P, Couderc E, Farinotti R, Henzel C, et al. Effects of diazepam and midazolam on baroreflex control of heart rate and on sympathetic activity in humans. Anesth Analg 1986;65:113-9. Submitted Jan 24, 2017; accepted May 24, 2017.