Abstract
Background
Although it is known that excessive intraoperative fluid and vasopressor agents are detrimental for anastomotic healing, optimal anesthesiology protocols for colorectal surgery are currently lacking.
Objective
To scrutinize the current hemodynamic practice and vasopressor use and their relation to colorectal anastomotic leakage.
Design
A secondary analysis of a previously published prospective observational study: the LekCheck study.
Study setting
Adult patients undergoing a colorectal resection with the creation of a primary anastomosis.
Outcome measures
Colorectal anastomotic leakage (CAL) within 30 days postoperatively, hospital length of stay and 30-day mortality.
Results
Of the 1548 patients, 579 (37%) received vasopressor agents during surgery. Of these, 201 were treated with solely noradrenaline, 349 were treated with phenylephrine, and 29 received ephedrine. CAL rate significantly differed between the patients receiving vasopressor agents during surgery compared to patients without (11.8% vs 6.3%, p < 0.001). CAL was significantly higher in the group receiving phenylephrine compared to noradrenaline (14.3% vs 6%, p < 0.001). Vasopressor agents were used more often in patients treated with Goal Directed Therapy (47% vs 34.6%, p < 0.001). There was a higher mortality rate in patients with vasopressors compared to the group without (2.8% vs 0.4%, p = 0.01, OR 3.8). Mortality was higher in the noradrenaline group compared to the phenylephrine and those without vasopressors (5% vs. 0.4% and 1.7%, respectively, p < 0.001). In multivariable analysis, patients with intraoperative vasopressor agents had an increased risk to develop CAL (OR 2.1, CI 1.3–3.2, p = 0.001).
Conclusion
The present study contributes to the evidence that intraoperative use of vasopressor agents is associated with a higher rate of CAL. This study helps to create awareness on the (necessity to) use of vasopressor agents in colorectal surgery patients in striving for successful anastomotic wound healing. Future research will be required to balance vasopressor agent dosage in view of colorectal anastomotic leakage.
Keywords: Colorectal anastomotic leakage, Vasopressors, Fluid management
Although research has shown that both excessive and restrictive intraoperative fluid administration as well as vasopressor use could possibly be detrimental for colonic anastomotic healing (CAL) [1–4], optimal anesthesiology protocols for colorectal surgery are currently lacking. Multiple strategies have been reported: restrictive fluid therapy, also called near-zero fluid balance, consists of replacing only the fluid that is lost during surgery, while Goal Directed Therapy (GDT) refers to the hemodynamic optimization during surgery by titrating fluids, using vasopressors and/or inotropes to reach predefined hemodynamic goals to maintain sufficient perfusion [5]. The literature is equivocal on the superiority of either of the two regimens. Earlier studies have suggested that restrictive fluid therapy is associated with a shorter length of stay and better short-term outcomes compared to a liberal fluid regimen [4, 6]. Few studies have investigated GDT in colorectal surgery with CAL as endpoint with a meta-analysis not showing a benefit [7]. For vasopressor use, the fourth updated Enhanced Recovery After Surgery (ERAS) Society guideline recommends intraoperative vasopressor use, although based on studies that were not colorectal surgery specific [8–10]. Therefore, this study aimed to investigate the current practice of intraoperative hemodynamic care in colorectal patients, including the current fluid management and vasopressor use and their relation to CAL.
Methods
Study design and population
From January 2016 to December 2019, 14 hospitals (11 in the Netherlands, one in Belgium, one in Italy and one in Australia) participated in the LekCheck study, a multicenter prospective cohort study [11]. In the LekCheck study, consecutive adult patients undergoing colorectal surgery with primary anastomosis were enrolled. A detailed description of the study design and the main results have been published previously. Informed consent was obtained from all patients. The administration of vasopressor agents was registered (yes/no) during the creation of the anastomosis intraoperatively. If at that moment the patient was receiving vasopressors, the answer yes was given and the type of drugs and dosage was noted. For the current analysis, patients treated with ephedrine were excluded from the analysis. In the noradrenaline patient group, the patients were solely treated with noradrenaline during surgery.
Data collection
The following variables were collected preoperatively: age, sex, diabetes mellitus (yes/no), body mass index (BMI), steroid use, intoxications (smoking status and alcohol), American Society of Anesthesiologist (ASA) score, benign or malignant disease as the indication for surgery, tumor node and metastasis (TNM) and American Joint Committee on Cancer (AJCC) stage, neoadjuvant therapy, tumor distance from the anal verge (AV) and preoperative hemoglobin level (HB). Intraoperatively the following parameters were collected: blood glucose level, use of epidural anesthesia, use and type and dosage of vasopressors, volume of blood loss, fluid administration, body temperature in degrees Celsius, mean arterial pressure (MAP), oxygen saturation, the occurrence of an intraoperative event (e.g., hypoxic events, hypertension, hypercarbia, bradycardia, hypotension, embolism, reanimation, formation of a stoma and stoma type, more extensive resection than planned, serosa lesions, bladder and ureteral injuries, intraoperative bleeding, splenectomy) and the assessment of fecal contamination. Postoperatively, the following variables were collected: the occurrence of CAL within 30 days, length of hospital stay and mortality.
Outcomes
Primary outcome was CAL. CAL was defined according to Reisinger, et al.: “Clinically relevant anastomotic leakage is defined as extra luminal presence of contrast fluid on contrast-enhanced CT scans and/or leakage when relaparotomy was performed, requiring re-intervention or treatment.” [12, 13]. Secondary outcomes were demise of the patient during hospital stay and length of hospital stay.
Missing data
Missing data were imputed using predictive mean matching with 10 iterations. Variables with more than 30% missing data were excluded.
Cutoff values and statistics
Patient characteristics and intraoperative variables between patients with and without vasopressors were compared. Subsequently, within the vasopressor group, patients who received noradrenaline were compared to patients with phenylephrine. Dichotomous and categorical data were expressed in percentages and frequencies. Continuous variables were presented as means or medians. Dichotomous and categorical data were expressed in percentages and frequencies. Continuous intraoperative parameters were prior to inclusion dichotomized to create a composite score. The preferable cutoff values were derived from a previously published review [6]. Only exceptions were a low hemoglobin value below 7 mmol/L, hyperglycemia defined as a glucose level of > 10 mmol/L and blood loss defined as > 500 mL during the procedure, making these cutoff values as unnegotiable risk factors. Dichotomous data were compared using the X2 test. Continuous data were compared using the Student’s t test or the Mann–Whitney U test (skewed distribution). The primary and secondary outcomes were analyzed in a univariate analysis with the administration of vasopressor agents as the independent and the outcome as the dependent variable. Then, a multivariable analysis was performed to adjust for confounders. Relevant confounders were defined as preoperative or intraoperative variables which altered the B value of the logistic regression model with more than 10%. A p value < 0.05 was regarded as statistically significant. Results are reported as odds ratios (OR) and 95% confidence intervals (CIs). Data were analyzed with Statistical Package for the Social Sciences software (SPSS, version 26).
Results
Data from 1821 colorectal patients were collected, 180 non-elective patients were excluded and another 93 due were excluded to incomplete data. A total of 1548 patients were found eligible. The median age was 69 years (range 21–95 years) and 803 (52%) of the patients were male. Of the 1548 patients, 579 (37%) received vasopressor agents during surgery. From the 579 patients in the vasopressor agents group, 201 were treated with solely noradrenaline, 349 were treated with phenylephrine, and 29 received ephedrine. There were 129 (8.3%) CAL cases.
Patient characteristics and intraoperative factors
Patients with vasopressors during surgery were more often older than 70 years (56% vs. 45%, p < 0.001), had more frequently diabetes mellitus (19% vs. 13%, p = 0.001), had more frequently a colorectal malignant diagnosis (86% vs. 80%, p = 0.003) and less often a tumor within 15 cm from the AV (27% vs. 36%, p = 0.016). Significant intraoperative factors differing between patients with and without vasopressor agents were epidural analgesia (37% vs. 31%, p = 0.010), GDT (26% vs. 16%, p < 0.001), blood loss > 500 mL (7% vs. 2%, p = 0.003), blood transfusion (4.5% vs. 1.1%, p < 0.001), low hemoglobin (21% vs 12%, p < 0.001) and fluid administration < 500 mL/h (41% vs. 48%, p = 0.036). See Tables 1 and 2.
Table 1.
Baseline characteristics of vasopressor agents (n = 1548)
| Variable | Vasopressor agents (n = 579) | No vasopressor agents (n = 969) | p value | ||
|---|---|---|---|---|---|
| Missing | Missing | ||||
| Sex (male) | 290 (50%) | 513 (53%) | 0.150 | ||
| Age (years) * | 72 (24–92) | 68 (21–95) | < 0.001 | ||
| ≥ 70 | 319 (56%) | 438 (45%) | |||
| Body mass index ≥ 30 kg/m2 | 128 (23%) | 190 (20%) | 0.113 | ||
| ASA classification | < 0.001 | ||||
| ≥ 3 | 197 (35%) | 187 (19%) | |||
| Diabetes mellitus | 107 (19%) | n = 7 | 123 (13%) | n = 5 | 0.001 |
| Intoxications | |||||
| Current smoker | 62 (12%) | n = 19 | 124 (13%) | n = 45 | 0.160 |
| Pack years ≥ 15 years | 117 (27%) | n = 23 | 224 (28%) | n = 43 | 0.435 |
| Alcohol intake ≥ 3 units/day | 27 (5%) | n = 21 | 45 (5%) | n = 35 | 0.511 |
| Steroid use (excl. inhalers) | 15 (2%) | n = 9 | 25 (3%) | n = 5 | 0.542 |
| Disease | n = 6 | 0.003 | |||
| Malignant | 491 (86%) | 774 (80%) | |||
| Benign | 83 (14%) | 194 (20%) | |||
| Neoadjuvant therapy | n = 20 | n = 47 | 0.057 | ||
| None | 494 (89%) | 795 (86%) | |||
| 5 × 5 radiotherapy | 30 (6%) | 72 (8%) | |||
| Chemotherapy | 14 (3%) | 19 (2%) | |||
| Chemoradiotherapy | 11 (2%) | 36 (4%) | |||
| Distance of tumor from AV < 15 cm | 88 (27%) | 213 (36%) | 0.004 | ||
| Pathological TNM stage | n = 7 | n = 7 | 0.458 | ||
| I (T1-2N0M0) | 217 (37%) | 339 (35%) | |||
| II (T3-4N0M0) | 119 (21%) | 169 (18%) | |||
| III (T1-4N1-2M0) | 127 (22%) | 203 (21%) | |||
| IV (T1-4N1-2M1) | 21 (4%) | 56 (6%) | |||
Bold values indicate significance of p value (p < 0.005)
Data are presented as number (%) unless stated otherwise
ASA American Society of Anesthesia score, TNM Tumor, node and metastasis classification, AV Anal verge
A p < 0.05 was considered statistically significant
*Data are presented as medians (range)
Table 2.
Surgery-related factors and risk for vasopressor agents
| Variable | Vasopressor agents (n = 579) | No vasopressor agents (n = 969) | p value | ||
|---|---|---|---|---|---|
| Missing | Missing | ||||
| Could vasopressor agents be stopped? | 322 (56%) | ||||
| Epidural | 209 (37%) | 291 (31%) | 0.010 | ||
| Goal Directed Therapy | 145 (26%) | 160 (16%) | < 0.001 | ||
| Intraoperative event* | 92 (16%) | 126 (13%) | 0.065 | ||
| Temperature 36 degrees | 109 (19%) | n = 1 | 195 (20%) | n = 7 | 0.272 |
| Glucose > 10 mmol/L | 52 (9%) | n = 13 | 76 (8%) | n = 49 | 0.299 |
| Antibiotics < 15 or > 60 min prior incision | 132 (24%) | n = 17 | 221 (24%) | n = 38 | 0.482 |
| Blood loss > 500 mL | 31 (7%) | n = 35 | 37 (2%) | n = 42 | 0.003 |
| Oxygen < 94% | 13 (22%) | 14 (15%) | 0.173 | ||
| Mean arterial pressure < 65 | 53 (9%) | 86 (9%) | 0.463 | ||
| Urine production < 30 mL/h | 98 (18%) | 156 (17%) | 0.350 | ||
| OR time > 180 min | 167 (29%) | 288 (30%) | 0.379 | ||
| Hemoglobin | 0.001 | ||||
| Men < 7 mmol/L | 48 (9%) | 42 (4%) | < 0.001 | ||
| Woman < 6.5 mmol/L | 111 (21%) | 106 (12%) | |||
| Surgical approach | 0.113 | ||||
| Open | 95 (16%) | 136 (14%) | |||
| Laparoscopy | 451 (78%) | 779 (80%) | |||
| Laparoscopy with conversion | 31 (5%) | 53 (5%) | 0.223 | ||
| Fluid administration | 0.169 | ||||
| < 250 mL/h | 89 (15%) | 120 (12%) | |||
| 250–1000 mL/h | 453 (78%) | 773 (80) | |||
| > 1000 mL/h | 37 (6%) | 76 (8%) | |||
| Fluid administration > 500 mL/h | 236 (41%) | 465 (48%) | 0.003 | ||
| Blood transfusion | < 0.001 | ||||
| 1 packed cells | 21 (4%) | 10 (1%) | |||
| 2 packed cells | 3 (0.5%) | 1 (0.1%) | |||
Bold values indicate significance of p value (p < 0.005)
Data are presented as number (%) or as medians (range) for categorical and continuous variables, respectively. N is number of inclusions if due to missing data this deviates from total
A p < 0.05 was considered statistically significant
*Intraoperative events include: hypoxic events, hypertension, hypercarbia, bradycardia, hypotension, embolism, reanimation, more extensive resection than planned, serosa lesions, bladder and ureteral injuries, intraoperative bleeding, splenectomy or bleeding
Table 3 shows the different characteristics between patients with intraoperative administration of noradrenaline or phenylephrine. Patients with noradrenaline more often had diabetes mellitus (24% vs. 15%, p = 0.006) and more often a tumor distance to the AV of < 15 cm (12% vs. 8%, p = 0.027). In univariate analysis, the intraoperative factors associated with noradrenaline patients were: use of an epidural (51% vs. 26%, p < 0.001), GDT (33% vs. 21%, p = 0.002), intraoperative event (21% vs. 13%, p = 0.027), temperature < 36.0 degrees Celsius (26% vs. 15%, p = 0.002), blood loss > 500 mL (10% vs. 5%, p = 0.037), mean arterial pressure (MAP) < 65 mmHg (12% vs. 7%, p = 0.029), operation time > 180 min (36% vs 26%, p = 0.010), conversion from laparoscopic to open surgery (9% vs. 4%, p < 0.001), fluid administration > 500 mL/h (48% vs. 67%, p < 0.001), high blood glucose > 10 mmol/L (16% vs. 6%, p < 0.001) and a blood transfusion given (7.5% vs. 2.6%, p = 0.048), see Table 4. GDT was followed in 19% (n = 305) of all patients. Vasopressors were used significantly more in these patients (47% in patients with GDT vs. 34.6% without, p < 0.001), see Table 5.
Table 3.
Baseline characteristics of noradrenaline or phenylephrine (n = 550)
| Variable | Noradrenaline (n = 201) | Phenylephrine (n = 349) | p value | ||
|---|---|---|---|---|---|
| Missing | Missing | ||||
| Sex (male) | 97 (48%) | 182 (52%) | 0.215 | ||
| Age (years)* | 71 (31–95) | 72 (24–94) | 0.455 | ||
| ≥ 70 | 115 (58%) | 193 (57%) | |||
| Body mass index ≥ 30 kg/m2 | 39 (20%) | 82 (24%) | 0.153 | ||
| ASA classification | 0.519 | ||||
| ≥ 3 | 70 (35%) | 121 (35%) | |||
| Diabetes mellitus | 48 (24%) | 51 (15%) | n = 6 | 0.006 | |
| Intoxications | |||||
| Current smoker | 23 (12%) | n = 10 | 34 (11%) | n = 23 | 0.348 |
| Pack years ≥ 15 years | 41 (32%) | n = 12 | 70 (25%) | n = 27 | 0.060 |
| Alcohol intake ≥ 3 units/day | 12 (6%) | n = 9 | 14 (4%) | n = 17 | 0.219 |
| Steroid use (excl. inhalers) | 6 (3%) | 9 (3%) | 0.504 | ||
| Disease | n = 5 | 0.423 | |||
| Malignant | 171 (85%) | 296 (86%) | |||
| Benign | 30 (15%) | 48 (14%) | |||
| Neoadjuvant therapy | n = 6 | n = 19 | 0.051 | ||
| None | 167 (86%) | 305 (93%) | |||
| 5 × 5 radiotherapy | 16 (8%) | 11 (3%) | |||
| Chemotherapy | 6 (3%) | 8 (2%) | |||
| Chemoradiotherapy | 5 (3%) | 4 (1%) | |||
| Distance of tumor from AV < 15 cm | 25 (12%) | 28 (8%) | 0.027 | ||
| Pathological TNM stage | n = 7 | n = 7 | 0.129 | ||
| I (T1-2N0M0) | 62 (31%) | 141 (40%) | |||
| II (T3-4N0M0) | 58 (29%) | 60 (17%) | |||
| III (T1-4N1-2M0) | 41 (20%) | 79 (23%) | |||
| IV (T1-4N1-2M1) | 6 (3%) | 13 (4%) | |||
Bold values indicate significance of p value (p < 0.005)
Data are presented as number (%) unless stated otherwise
ASA American Society of Anesthesia score, TNM Tumor, node and metastasis classification, AV Anal verge
A p < 0.05 was considered statistically significant
*Data are presented as medians (range)
Table 4.
Surgery-related factors and use noradrenaline or phenylephrine (n = 550)
| Variable | Noradrenaline (n = 201) | Phenylephrine (n = 349) | p value | ||
|---|---|---|---|---|---|
| Missing | Missing | ||||
| Could vasopressor agents be stopped? | 68 (34%) | 238 (68%) | < 0.001 | ||
| Epidural | 102 (51%) | 91 (26%) | < 0.001 | ||
| Goal directed therapy | 63 (33%) | 74 (21%) | 0.002 | ||
| Intraoperative event* | 41 (21%) | 48 (13%) | 0.027 | ||
| Temperature < 36 degrees | 51 (26%) | n = 1 | 53 (15%) | 0.002 | |
| Glucose > 10 mmol/L | 30 (16%) | n = 11 | 21 (6%) | n = 1 | < 0.001 |
| Antibiotics < 15 or > 60 min prior incision | 52 (27%) | n = 10 | 71 (21%) | n = 7 | 0.057 |
| Blood loss > 500 mL | 18 (10%) | n = 11 | 15 (5%) | n = 28 | 0.037 |
| Oxygen < 94% | 5 (3%) | 8 (2%) | 0.544 | ||
| Mean arterial pressure < 65 mmhg | 24 (12%) | 24 (7%) | 0.029 | ||
| Urine production < 30 mL/h | 19 (9%) | 78 (22%) | < 0.001 | ||
| OR time > 180 min | 73 (36%) | 92 (26%) | 0.010 | ||
| Hemoglobin | 0.284 | ||||
| Men < 6.5, woman < 6 mmol/L | 16 (8%) | 31 (10%) | |||
| < 7 mmol/L | 39 (19%) | 67 (21%) | 0.340 | ||
| Surgical approach | < 0.001 | ||||
| Open | 55 (16%) | 32(14%) | |||
| Laparoscopy | 127 (63%) | 303 (87%) | |||
| Laparoscopy with conversion | 18 (9%) | 13 (4%) | 0.223 | ||
| Fluid administration | 0.103 | ||||
| < 250 mL/h | 26 (13%) | 62 (18%) | |||
| 250–1000 mL/h | 158 (79%) | 270 (77%) | |||
| > 1000 mL/h | 17 (8%) | 17 (5%) | |||
| Fluid administration > 500 mL/h | 97 (48%) | 232 (67%) | < 0.001 | ||
| Blood transfusion | 0.048 | ||||
| 1 packed cells | 13 (7%) | 8 (2%) | |||
| 2 packed cells | 1 (0.5%) | 2 (0.6%) | |||
Bold values indicate significance of p value (p < 0.005)
Data are presented as number (%) or as medians (range) for categorical and continuous variables, respectively. N is number of inclusions if due to missing data this deviates from total
*Intraoperative events include: hypoxic events, hypertension, hypercarbia, bradycardia, hypotension, embolism, reanimation, more extensive resection than planned, serosa lesions, bladder and ureteral injuries, intraoperative bleeding, splenectomy or bleeding. A p < 0.05 was considered statistically significant
Table 5.
Univariate analysis of Goal Directed Therapy
| GDT (n = 305) | No GDT (n = 1221) | p value | |
|---|---|---|---|
| Vasopressor | 145 (47.5%) | 422 (34.6%) | < 0.001 |
| Urine output 30 mL/h | 53 (18%) | 201 (18%) | 0.530 |
| Blood loss > 500 mL | 16 (5.9%) | 51 (4.6%) | 0.238 |
| Noradrenaline | 63 (20.6%) | 127 (10.4%) | |
| Phenylephrine | 74 (24.3%) | 275 (22.5%) | |
| Fluid < 250 mL/h | 46 | 161 | |
| Fluid > 250– < 500 mL/h | 234 | 977 | |
| Fluid > 1000 mL/h | 25 | 83 |
In the vasopressor agents group, there were 68 CAL versus 61 CAL in the group without vasopressors (11.7% vs 6.3%, p < 0.001). In the phenylephrine group, there were 50 CAL and in the noradrenaline group 18 CAL (14.3% vs 6%, p = 0.002). Multivariable analysis showed that intraoperative vasopressor agents had an OR of 2.1 to develop CAL (CI 1.3–3.2), see Table 6. Multivariable analysis also showed that phenylephrine had a higher risk of developing CAL compared to noradrenaline (OR 4.2; CI 1.9–8.6), see Table 7. There were no significant differences between fluid administration and vasopressive agents in colon (p = 0.73) or rectal surgery (p = 0.45) in multivariate analysis. Only in malignant resections, multivariable analysis showed that the odds of receiving vasopressors during surgery is twice as high versus benign surgery (OR 2.0, CI 1.2–3.3, p = 0.004). Open surgery is not independently associated with a higher risk of CAL in multivariate analysis. However, a multivariate analysis found that open versus laparoscopic surgery reduced the chance of receiving less than 500 mL/h of fluids (OR 0.5, CI 0.3–0.8, p = 0.010).
Table 6.
Multivariate analysis of postoperative outcomes with vasopressor agents and non-vasopressors agents
| Vasopressor (579) | No Vasopressor (971) | Total (n = 1549) | Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | p value | ||||
| Anastomotic leakage | 68 (11.8%) | 61 (6.3%) | 129 (8.3%) | 2.0 (1.3–2.8) | < 0.001 | 2.1 (1.3–3.2)* | 0.001* |
| Mortality | 16 (2.8%) | 4 (0.4%) | 129 (8.3%) | 6.9 (2.3–21) | 0.001 | 5.1 (1.6–16)** | 0.005** |
| Length of stay | M = 5 | M = 4 | M = 4 | 1.5 (1.2–1.9) | 0.001 | 1.2 (0.9–1.5)*** | 0.187 |
OR Odds ratio, CI Confidence interval, M Median
*In the multivariate analyses adjusted for epidural, pack years > 15 and low hemoglobin
**In the multivariate analyses adjusted for pack years > 15 and low hemoglobin
***In the multivariate analyses adjusted for ASA > 3 and low hemoglobin
Table 7.
Multivariate analysis of postoperative outcomes with noradrenaline and phenylephrine group
| Noradrenaline (n = 201) | Phenylephrine (n = 349) | Total (n = 1549) | Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|---|---|---|
| OR (95% CI) | p value | OR (95% CI) | p value | ||||
| Anastomotic leakage | 68 (11.8%) | 61 (6.3%) | 129 (8.3%) | 2.6 (1.4–5.1) | 0.004 | 4.2 (1.9–8.6)* | < 0.001* |
| Mortality | 10 (5%) | 6 (1.7%) | 129 (8.3%) | 0.3 (0.1–0.9) | 0.037 | 0.3 (0.1–0.8)** | 0.019** |
| Length of stay | M = 6 | M = 4 | M = 4 | 0.6 (0.4–0.9) | 0.024 | 0.6 (0.3–1.1)*** | 0.052*** |
OR Odds ratio, CI Confidence interval, M Median
*In the multivariate analyses adjusted for temperature < 36 degrees, low hemoglobin and OR time > 180 min
**In the multivariate analyses adjusted for ASA > 3
***In the multivariate analyses adjusted for low hemoglobin, pack years > 15, intraoperative event and epidural
There was a 1.3% mortality rate in the total study population. Sixteen of these 20 patients received vasopressor agents during surgery (2.8% vs 0.4%, p < 0.001). Within the vasopressor agents group, 30-day mortality rate was 5% in the noradrenaline group compared to 1.7% in the phenylephrine group (p = 0.002). Multivariable analysis showed that the use of vasopressors had an OR of 5.1 for mortality compared to patients without intraoperative vasopressor agent use (CI 1.6–16), see Table 6.
The median length of hospital stay was not different for patients with or without vasopressors. There was no significant association in the occurrence of CAL between patients treated with or without GDT.
Discussion
This study scrutinized the current practice of intraoperative hemodynamic care in colorectal patients. The prospectively collected data showed that the use of intraoperative vasopressor agents was related to a significantly higher CAL and mortality rate.
This could be due to the physiology of vasopressors during a procedure not being beneficial for the healing of the anastomosis. The factors that showed a relation to vasopressor agent use are: age > 70 years, diabetes mellitus, malignant diseases, distance of tumor to AV < 15 cm, epidural use, GDT, blood loss > 500 mL, blood transfusion, low hemoglobin and fluid administration < 500 mL/h. The question is whether the vasopressor agents themselves exert a detrimental effect on CAL and survival, or that this effect should be contributed to the circumstances that led to the decision to administer vasopressor agents.
The results of this study are compliant with those of the study by Adanir et al. [2], who showed that vasopressors appeared to increase the risk of CAL due to vasoconstriction, deterioration of microcirculation and possibly local hypoxia. Similarly, Choudhuri et al. [3] found that patients who required inotropic support during surgery had a four times higher risk of CAL.
In the RELIEF study [14], the largest trial to date of perioperative fluid management, restrictive fluid management was associated with a higher rate of acute kidney injury. Myles et al. showed that restrictive fluid regimen was not associated with a higher rate of disability-free survival versus liberal fluid management [15]. Over the past decade, patients undergoing colorectal surgery were increasingly subject to a restrictive fluid management [16–18]. Studies showed that a liberal fluid regime could lead to an increase in interstitial volume, cardiopulmonary dysfunction, inflammation, edema and impaired tissue oxygenation. This is due to hypervolemia causing shedding of the endothelial glycocalyx and therefore affecting the vascular permeability [19]. An optimal threshold for stroke volume variation to prevent this still needs to be determined [20]. Anastomotic wound healing is adversely affected by intestinal edema [17]. Consequently, The American Society for Enhanced Recovery and Perioperative Quality Initiative created a framework for perioperative fluid management and concluded that GDT is safe in the majority of colorectal patients [18, 21]. Correct patient allocation for GDT, however, is still under debate. GDT could be the solution for not giving too much fluid during surgery, but the benefits in outcome have yet to be proven [21, 22]. The current study found that in patients treated with GDT more vasopressors were used. It could be that the avoidance of excessive fluid turned into an inordinate restrictive regime [23]. An other possibility is that GDT is more often applied in higher risk patients. Recent studies show that a restrictive fluid management and GDT also have their drawbacks [21, 22], possibly because more vasopressors are used to compensate for blood pressure and heart rate.
There are some limitations to this study. Despite our multivariate analysis, in which was tried to correct for patient characteristics and intraoperative variables, a firm conclusion cannot be drawn due to an unclear cause–effect relation. For instance, pre-existing risk factors, such as comorbidities and higher ASA-scores, could in itself be a cause for more vasopressor agents during surgery, while it is also known that these patients have a higher risk for developing CAL [24]. The administration of vasopressors may therefore rather be an intraoperative requirement and reflect the patient’s preoperative condition [25]. The lack of postoperative information limits the ability to draw conclusions between the outcomes and the use of vasopressors and fluid restriction postoperatively. Nevertheless, this study shows an association with CAL and mortality in patients with certain vasopressor agents. Currently, our group is mapping current anesthesia practice for colorectal surgery.
In the future, a prospective follow-up study that stratifies for intraoperative conditions could contribute to the knowledge regarding the association between specific vasopressive agents and outcomes such as CAL and mortality.
The present study contributes to the evidence that intraoperative use of vasopressor agents is associated with a higher rate of CAL. This study helps to create awareness on the (necessity to) use of vasopressor agents in colorectal surgery patients in striving for successful anastomotic wound healing. Future research will be required to balance vasopressor agent dosage in view of colorectal anastomotic leakage.
Acknowledgements
LekCheck Study group: Rudi M.H. Roumen MD PhD, Department of Anesthesiology, Amsterdam University Medical Centers, Location VUmc, Amsterdam, The Netherlands; Stefanus J. van Rooijen MD PhD, Care and Public Health Research Institute, Maastricht University, Maastricht, The Netherlands; Wim Bleeker MD, Department of Surgery, Wilhelmina Ziekenhuis, Assen, The Netherlands; Laurents P.S. Stassen Prof Dr MD PhD, Department of Surgery, Maastricht University Medical Centre, Maastricht, The Netherlands; Audrey Jongen MD PhD, Department of Surgery, Maastricht University Medical Centre, Maastricht, The Netherlands; Carlo V. Feo MD FACS, Department of Surgery, Ospedale del Delta, Lagosanto, Ferrara, Italy; Simone Targa MD, Department of Surgery, Ospedale del Delta, Lagosanto, Ferrara, Italy; Niels Komen MD PhD, Department of Surgery, Antwerp University Hospital, Antwerp, Belgium; Hidde M. Kroon MD PhD, Colorectal Unit, Department of Surgery, Royal Adelaide Hospital, University of Adelaide, Adelaide, Australia; Tarik Sammour MBCHB PhD FRACS, Colorectal Unit, Department of Surgery, Royal Adelaide Hospital, University of Adelaide, Adelaide, Australia; Emmanuel A.G.L. Lagae MD, Department of Surgery, ZorgSaam, Terneuzen, The Netherlands; Aalbert K. Talsma MD, Department of Surgery, Deventer Ziekenhuis, Deventer, The Netherlands; Johannes A. Wegdam MD, Department of Surgery, Elkerliek Ziekenhuis, Helmond, The Netherlands; Tammo S. de Vries Reilingh MD, Department of Surgery, Elkerliek Ziekenhuis, Helmond, The Netherlands; Bob van Wely MD, Department of Surgery, Bernhoven, Uden, the Netherlands; Marie J. van Hoogstraten MD, Department of Surgery, Bernhoven, Uden, the Netherlands; Dirk J.A. Sonneveld MD, Department of Surgery, Dijklander Ziekenhuis, Hoorn, The Netherlands; Emiel G.G. Verdaasdonk MD, Department of Surgery, Jeroen Bosch Ziekenhuis, Den Bosch, The Netherlands.
Abbreviations
- AL
Anastomotic leakage
- BMI
Body mass index
- ASA
American Society of Anesthesiologists
- AJCC
American Joint Committee on Cancer
- MAP
Mean arterial pressure
- TNM
Tumor node and metastasis
- ERAS
Enhanced recovery after surgery
Funding
This work received support from European Association for Endoscopic Surgery and other Interventional Techniques (2019 Research Grant).
Declarations
Disclosures
DEH is employed by Johnson and Johnson, although her work is not related to the subject of this paper. FD was granted the 2019 EAES research grant for this project. He received a speaker fee from Medtronic. BB, EWI, MR, GDS and JS have no conflicts of interest or financial ties to disclose.
Footnotes
The members of LekCheck Study group are listed in the acknowledgement section.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Daitlin E. Huisman, Email: d.huisman@amsterdamumc.nl
the LekCheck Study group:
Rudi M. H. M. H. Roumen, Stefanus J. van Rooijen, Wim Bleeker, Laurents P. S. Stassen, Audrey Jongen, Carlo V. Feo, Simone Targa, Niels Komen, Hidde M. Kroon, Tarik Sammour, Emmanuel A. G. L. Lagae, Aalbert K. Talsma, Johannes A. Wegdam, Tammo S. de Vries Reilingh, Bob van Wely, Marie J. van Hoogstraten, Dirk J. A. Sonneveld, and Emiel G. G. Verdaasdonk
References
- 1.Zakrison T, Nascimento BA, Tremblay LN, Kiss A, Rizoli SB. Perioperative vasopressors are associated with an increased risk of gastrointestinal anastomotic leakage. World J Surg. 2007;31(8):1627–1634. doi: 10.1007/s00268-007-9113-4. [DOI] [PubMed] [Google Scholar]
- 2.Adanir T, Nazli O, Kara C, Aksun M, Sozutek A, Sencan A, Oktay G. The relationship between vasopressor dose and anastomotic leak in colon surgery: an experimental trial. Int J Surg. 2010;8(3):221–224. doi: 10.1016/j.ijsu.2010.01.004. [DOI] [PubMed] [Google Scholar]
- 3.Choudhuri AH, Uppal R, Kumar M. Influence of non-surgical risk factors on anastomotic leakage after major gastrointestinal surgery: audit from a tertiary care teaching institute. Int J Crit Illn Inj Sci. 2013;3(4):246–249. doi: 10.4103/2229-5151.124117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Brandstrup B, Tønnesen H, Beier-Holgersen R, Hjortsø E, Ørding H, Lindorff-Larsen K, Rasmussen MS, Lanng C, Wallin L, Iversen LH, Gramkow CS, Okholm M, Blemmer T, Svendsen PE, Rottensten HH, Thage B, Riis J, Jeppesen IS, Teilum D, Christensen AM, Graungaard B, Pott F, Danish Study Group on Perioperative Fluid Therapy Effects of intravenous fluid restriction on postoperative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg. 2003;238(5):641–648. doi: 10.1097/01.sla.0000094387.50865.23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cove ME, Pinsky MR. Perioperative hemodynamic monitoring. Best Pract Res Clin Anaesthesiol. 2012;26:453–462. doi: 10.1016/j.bpa.2012.10.003. [DOI] [PubMed] [Google Scholar]
- 6.van Rooijen SJ, Huisman D, Stuijvenberg M, Stens J, Roumen RMH, Daams F, Slooter GD. Intraoperative modifiable risk factors of colorectal anastomotic leakage: why surgeons and anesthesiologists should act together. Int J Surg. 2016;36(Part A):183–200. doi: 10.1016/j.ijsu.2016.09.098. [DOI] [PubMed] [Google Scholar]
- 7.Rollins KE, Lobo DN. Intraoperative goal-directed fluid therapy in elective major abdominal surgery: a meta-analysis of randomized controlled trials. Ann Surg. 2016;263:465–476. doi: 10.1097/SLA.0000000000001366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gustafsson UO, Scott MJ, Hubner M, Nygren J, Demartines N, Francis N, et al. Guidelines for perioperative care in elective colorectal surgery: enhanced recovery after surgery (ERAS((R))) society recommendations: 2018. World J Surg. 2019;43(3):659–695. doi: 10.1007/s00268-018-4844-y. [DOI] [PubMed] [Google Scholar]
- 9.Feldheiser A, Conroy P, Bonomo T, et al. Development and feasibility study of an algorithm for intraoperative goal directed haemodynamic management in noncardiac surgery. J Int Med Res. 2012;40:1227–1241. doi: 10.1177/147323001204000402. [DOI] [PubMed] [Google Scholar]
- 10.Bijker JB, van Klei WA, Vergouwe Y, et al. Intraoperative hypotension and 1-year mortality after noncardiac surgery. Anesthesiology. 2009;111:1217–1226. doi: 10.1097/ALN.0b013e3181c14930. [DOI] [PubMed] [Google Scholar]
- 11.Huisman DE, Reudink M, van Rooijen SJ, Bootsma BT, van de Brug T, Stens J, Bleeker W, Stassen LPS, Jongen A, Feo CV, Targa S, Komen N, Kroon HM, Sammour T, Lagae EAGL, Talsma AK, Wegdam JA, de Vries Reilingh TS, van Wely B, van Hoogstraten MJ, Sonneveld DJA, Veltkamp SC, Verdaasdonk EGG, Roumen RMH, Slooter GD, Daams F. LekCheck: a prospective study to identify perioperative modifiable risk factors for anastomotic leakage in colorectal surgery. Ann Surg. 2022;275(1):e189–e197. doi: 10.1097/SLA.0000000000003853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Reisinger KW, Poeze M, Hulsewe KW, van Acker BA, van Bijnen AA, Hoofwijk AG, et al. Accurate prediction of anastomotic leakage after colorectal surgery using plasma markers for intestinal damage and inflammation. J Am Coll Surg. 2014;219(4):744–751. doi: 10.1016/j.jamcollsurg.2014.06.011. [DOI] [PubMed] [Google Scholar]
- 13.MacKay G, Fearon K, McConnachie A. Randomized clinical trial of the effect of postoperative intravenous fluid restriction on recovery after elective colorectal surgery. Br J Surg. 2006;93:1469–1474. doi: 10.1002/bjs.5593. [DOI] [PubMed] [Google Scholar]
- 14.Miller TE, Pearse RM. Perioperative fluid management: moving toward more answers than questions-a commentary on the RELIEF study. Perioper Med (London, England) 2019;8:2. doi: 10.1186/s13741-019-0113-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Myles PS, Bellomo R, Corcoran T, Forbes A, Peyton P, Story D, et al. Restrictive versus liberal fluid therapy for major abdominal surgery. N Engl J Med. 2018;378(24):2263–2274. doi: 10.1056/NEJMoa1801601. [DOI] [PubMed] [Google Scholar]
- 16.Abraham-Nordling M, Hjern F, Pollack J. Randomized clinical trial of fluid restriction in colorectal surgery. Br J Surg. 2012;99:186–191. doi: 10.1002/bjs.7702. [DOI] [PubMed] [Google Scholar]
- 17.Chestovich PJ, Lin AY, Yoo J. Fast-track pathways in colorectal surgery. Surg Clin N Am. 2013;93:21–32. doi: 10.1016/j.suc.2012.09.003. [DOI] [PubMed] [Google Scholar]
- 18.Thiele RH, Raghunathan K, Brudney CS, Lobo DN, Martin D, Senagore A, Cannesson M, Gan TJ, Mythen MM, Shaw AD, Miller TE, Perioperative Quality Initiative (POQI) I Workgroup American Society for Enhanced Recovery (ASER) and Perioperative Quality Initiative (POQI) joint consensus statement on perioperative fluid management within an enhanced recovery pathway for colorectal surgery. Perioper Med (London, England) 2016;5:24. doi: 10.1186/s13741-016-0049-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chappell D, Bruegger D, Potzel J, Jacob M, Brettner F, Vogeser M, Conzen P, Becker BF, Rehm M. Hypervolemia increases release of atrial natriuretic peptide and shedding of the endothelial glycocalyx. Crit Care. 2014;18(5):538. doi: 10.1186/s13054-014-0538-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Liu Y, Chen G, Gao J, Chi M, Mao M, Shi Y, Ji Z, Fu Q, Zhang H, Xu Z. Effect of different levels of stroke volume variation on the endothelial glycocalyx of patients undergoing colorectal surgery: a randomized clinical trial. Exp Physiol. 2021;106(10):2124–2132. doi: 10.1113/EP089348. [DOI] [PubMed] [Google Scholar]
- 21.Kaufmann T, Clement RP, Scheeren TW, Saugel B, Keus F, van der Horst IC. Perioperative goal-directed therapy: a systematic review without meta-analysis. Acta Anaesthesiol Scand. 2018;62(10):1340–1355. doi: 10.1111/aas.13212. [DOI] [PubMed] [Google Scholar]
- 22.Makaryus R, Miller TE, Gan TJ. Current concepts of fluid management in enhanced recovery pathways. Br J Anaesth. 2018;120(2):376–383. doi: 10.1016/j.bja.2017.10.011. [DOI] [PubMed] [Google Scholar]
- 23.Lilot M, Ehrenfeld JM, Lee C, Harrington B, Cannesson M, Rinehart J. Variability in practice and factors predictive of total crystalloid administration during abdominal surgery: retrospective two-centre analysis. Br J Anaesth. 2015;114(5):767–776. doi: 10.1093/bja/aeu452. [DOI] [PubMed] [Google Scholar]
- 24.Dekker JWT, Liefers GJ, de Mol van Otterloo JCA, Putter H, Tollenaar RAEM. Predicting the risk of anastomotic leakage in left-sided colorectal surgery using a colon leakage score. J Surg Res. 2011;166(1):e27–e34. doi: 10.1016/j.jss.2010.11.004. [DOI] [PubMed] [Google Scholar]
- 25.Malbrain M, Langer T, Annane D, Gattinoni L, Elbers P, Hahn RG, et al. Intravenous fluid therapy in the perioperative and critical care setting: executive summary of the International Fluid Academy (IFA) Ann Intensive Care. 2020;10(1):64. doi: 10.1186/s13613-020-00679-3. [DOI] [PMC free article] [PubMed] [Google Scholar]