Abstract
Background.
Hepatic artery infusion pump (HAIP) chemotherapy is an advanced cancer therapy for primary and secondary hepatic malignancies. The risk of concurrent hepatic and/or colorectal operations with HAIP placement is unknown. Our objective was to characterize the short-term outcomes of concurrent surgery with HAIP placement.
Methods.
The 2005–2017 ACS NSQIP dataset was queried for patients undergoing hepatic and colorectal operations with or without HAIP placement. Outcomes were compared for HAIP placement with different combined procedures. Patients who underwent procedures without HAIP placement were propensity score matched with those with HAIP placement. The primary outcome was 30-day death or serious morbidity (DSM). Secondary outcomes included infectious complications, wound complications, length of stay (LOS), and operative time.
Results.
Of 467 patients who underwent HAIP placement, 83.9% had concurrent surgery. The rate of DSM was 10.7% for HAIP placement alone, 19.2% with concurrent minor hepatic procedures, 22.1% with concurrent colorectal resection, 23.2% with concurrent minor hepatic plus colorectal procedures, 28.4% with concurrent major hepatic resection, and 41.7% with concurrent major hepatic plus colorectal resection. On matched analyses, there was no difference in DSM, infectious, or wound complications for procedures with HAIP placement compared with the additional procedure alone, but operative time (294.7 vs 239.8 min, difference 54.9, 95% CI 42.8–67.0) and LOS (6 vs 5, IRR 1.20, 95% CI 1.08–1.33) were increased.
Conclusions.
HAIP placement is not associated with additional morbidity when performed with hepatic and/or colorectal surgery. Decisions regarding HAIP placement should consider the risks of concurrent operations, and patient- and disease-specific factors.
Management of metastatic cancer isolated to the liver (i.e., metastatic colorectal cancer) and unresectable primary hepatic malignancy (i.e., intrahepatic cholangiocarcinoma) is challenging and requires the coordinated effort of a multidisciplinary oncology team.1, 2 Long-term outcomes are aided by controlling or eradicating disease in the liver.3,4 One technique with demonstrated effectiveness in selected patients is hepatic artery infusion pump (HAIP) chemotherapy, which delivers high dose chemotherapy directly to the entire liver.5,6 For the past several decades, HAIP chemotherapy has been used to treat unresectable metastatic colorectal cancer isolated to the liver, as an adjuvant therapy following complete resection of metastatic colorectal cancer to the liver, and among patients with unresectable intrahepatic cholangiocarcinoma.7,8 However, the risks of this treatment are not insignificant, as HAIP chemotherapy requires a major operation to surgically place a catheter directly into the hepatic arterial system.
Because of limited expertise in the placement and management of HAIPs, few institutions include this therapy in their armamentarium. Recently, however, interest in HAIP chemotherapy has been increasing, with new centers offering the procedure and a multicenter adjuvant trial currently in process.9-11 Despite this, limited data exist informing the timing, sequence, and risks of HAIP placement. For many patients, such as those with unresectable intrahepatic cholangiocarcinoma, HAIP placement is performed as a standalone, definitive procedure. In other patients, such as those with metastatic colorectal cancer, HAIP placement is one component of a treatment plan that may include additional liver and/or colorectal procedures to eradicate all known disease.12 These procedures are often performed in a staged fashion to minimize surgical morbidity or as an attempt at conversion to resectability.
Concurrent procedures, particularly colorectal resection, have been included in prior HAIP placement studies and appear to be safe at experienced centers.13,14 However, studies assessing short-term outcomes of concurrent compared with individual procedures are limited.15 Although staged surgery may limit the perioperative risk of multiple major operations (e.g., right hepatic lobectomy, proctectomy, and HAIP placement), this must be weighed against the morbidity of multiple separate major abdominal procedures. Short-term outcomes data for different potential combinations of procedures would aid operative planning, clinical decision making, and patient counseling regarding the risks of HAIP placement with other procedures. Therefore, our objectives were to (1) characterize the short-term risks of HAIP placement with different combinations of synchronous surgery and (2) estimate the additional short-term risk when HAIP placement is performed concurrently with hepatic and/or colorectal procedures.
METHODS
Data Source
Patients were identified in the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) dataset from 2005 through 2017. ACS NSQIP employs a sampling methodology to select a subset of patients who have undergone surgery at participating hospitals. Data are clinically abstracted by trained surgical reviewers. The ACS NSQIP dataset has been previously described in detail and validated in the literature.16-20
Patient Population
Patients who underwent HAIP placement were identified by a primary or concurrent Current Procedure Terminology (CPT) code of 36260. These patients were assigned to cohorts based on whether they concurrently underwent colorectal resection, major hepatic resection (total lobectomy or trisegmentectomy), minor hepatic procedures (one or multiple partial lobectomies, wedge resection, or ablation procedures), concurrent major hepatic plus colorectal resection, concurrent minor hepatic plus colorectal resection, or no additional procedures (“Appendix 1”). Patients that underwent the aforementioned procedures without concurrent HAIP placement were included as a comparison group. Patients were excluded if the primary CPT and International Classification of Disease version 9 or 10 codes (ICD-9, ICD-10) suggested erroneous coding of a non-hepatic infusion device (i.e., pain-control pump, central venous port). Patients who underwent a concurrent major intra-abdominal surgery (e.g., gastrectomy or pancreatectomy) that did not fall within the defined cohorts were also excluded. This study was exempted from approval by the Northwestern University Institutional Review Board.
Primary and Secondary Outcomes
The primary outcome in this study was a binary variable composed of 30-day postoperative death and serious morbidity (DSM). This encompassed any occurrence of death, deep or organ space surgical site infection, wound dehiscence, pneumonia, prolonged mechanical ventilation > 48 h after surgery, unplanned reintubation, myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, stroke with neurologic deficit, deep vein thrombosis, pulmonary embolism, renal failure requiring dialysis, sepsis, septic shock, blood transfusion, or unplanned reoperation. This outcome has been previously used as a global measure of severe complications in studies employing NSQIP. 21-24 Secondary outcome measures included wound complications, consisting of any surgical site infection or wound dehiscence, and infectious complications, consisting of any surgical site infection, urinary tract infection, pneumonia, sepsis, or septic shock. Operative time and total hospital length of stay (LOS) were also included as secondary outcomes.
Statistical Analysis
Bivariate analyses comparing patient characteristics, primary, and secondary outcomes among procedural cohorts for patients who received a HAIP were performed using χ2 tests for categorical variables and one-way analysis of variance for continuous variables. Multivariable logistic regression models were constructed to evaluate the association between concurrent major surgery at the time of HAIP placement and DSM, wound complications, and infectious complications. A multivariable Poisson regression model was used to estimate the association with LOS and a multivariable linear regression model was used to estimate the association with operative time. Patient demographic factors, including age, sex, and race/ethnicity were included prospectively in the models. Additional clinical factors that differed among procedural cohorts or between cases and controls of the primary outcome of DSM at a significance level of α < 0.1 were also included.
To determine whether the addition of HAIP placement to hepatic and/or colorectal surgery was associated with an increased risk of complications, an appropriate control group of patients was identified using a propensity-score-matching algorithm. Potential matches were selected from patients who underwent hepatic and/or colorectal resection without HAIP placement. Patients with ages or primary diagnosis codes not represented among patients who underwent HAIP placement were excluded from the pool of potential matches. Propensity score matching was performed to match patients who underwent hepatic and/or colorectal resections without HAIP placement with those who underwent those procedures with HAIP placement. An exact match was required for procedural cohort (colorectal, major hepatic, minor hepatic, major hepatic plus colorectal, minor hepatic plus colorectal), cancer category (primary hepatobiliary, colorectal/metastasis, other/unknown) and presence of disseminated cancer. Additional covariates used in propensity score estimation included age, gender, body mass index, race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other/unknown), year of operation, smoking status prior to surgery, American Society of Anesthesiologists (ASA) physical status classification (I/II vs III/IV), number of medical comorbidities (0, 1, 2, or more), and surgical wound class. Matching was performed using greedy nearest neighbor matching with a 1:3 ratio of cases to controls.25
Patients undergoing concurrent surgery with HAIP placement were compared with the propensity-matched cohort without HAIP placement. Evaluation of match quality was performed by comparing predictor variables between the two groups using Fisher’s exact and χ2 tests for categorical variables and Student’s t test with equal variance for continuous variables. Conditional logistic regression models accounting for matching were constructed to assess association between HAIP placement and DSM, wound complications, and infectious complications. Conditional Poisson and linear regression models accounting for matching were used to evaluate for association with LOS and operative time, respectively.
For all statistical analyses, significance was calculated using two-sided tests with a threshold of α < 0.05. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
RESULTS
A total of 467 patients were identified who underwent HAIP placement between 2005 and 2017. Of these, 75 (16.1%) did not undergo concurrent hepatic and/or colorectal operations. Among the 392 patients who underwent concurrent procedures, 77 (16.5%) underwent concurrent colorectal resection, 67 (14.3%) underwent concurrent major hepatic resection, 167 (35.7%) underwent concurrent minor hepatic procedures, 12 (2.6%) underwent concurrent major hepatic plus colorectal resection, and 69 (14.3%) underwent concurrent minor hepatic plus colorectal procedures. At least one HAIP was placed at 37 hospitals over the 13-year period; however, in any given year, only 4 to 11 hospitals performed a HAIP placement. The majority of hospitals performed few procedures, with 5 hospitals in our sample placing more than 10 HAIPs. Patient cohorts varied in age and by primary cancer site, ASA classification, wound classification, and presence of disseminated cancer (Table 1).
TABLE 1.
Patient demographics and characteristics, by type of concurrent surgery performed at the time of HAIP placement
| HAIP alone (n = 75) |
Concurrent colorectal (n = 77) |
Concurrent major hepatic (n = 67) |
Concurrent minor hepatic (n = 167) |
Concurrent major hepatic and colorectal (n = 12) |
Concurrent minor hepatic and colorectal (n = 69) |
P value | |
|---|---|---|---|---|---|---|---|
| Mean age (SD) | 57.6 (11.9) | 54.5 (10.6) | 53.7 (12.0) | 55.2 (11.5) | 55.8 (11.0) | 50.5 (9.9) | 0.010 |
| Sex | 0.437 | ||||||
| Male (%) | 41 (54.7) | 51 (66.2) | 34 (50.8) | 102 (61.1) | 6 (50.0) | 41 (59.4) | |
| Female (%) | 34 (45.3) | 26 (33.8) | 33 (49.3) | 65 (38.9) | 6 (50.0) | 28 (40.6) | |
| Mean body mass index (SD) | 27.0 (5.4) | 28.0 (5.6) | 27.5 (5.4) | 28.9 (5.9) | 28.6 (4.4) | 27.6 (6.6) | 0.195 |
| Race | 0.649 | ||||||
| Non-Hispanic White (%) | 61 (81.3) | 56 (72.7) | 53 (79.1) | 135 (80.8) | 8 (66.7) | 50 (72.5) | |
| Non-Hispanic Black (%) | 4 (5.3) | 5 (6.5) | 3 (4.5) | 7 (4.2) | 1 (8.3) | 1 (1.5) | |
| Hispanic (%) | 3 (4.0) | 2 (2.6) | 4 (6.0) | 4 (2.4) | 1 (8.3) | 3 (4.4) | |
| Other/unknown (%) | 7 (9.3) | 14 (18.2) | 7 (10.5) | 21 (12.6) | 2 (16.7) | 15 (21.7) | |
| Cancer site | < 0.001 | ||||||
| Primary liver malignancy (%) | 12 (16.0) | 0 (0) | 5 (7.5) | 14 (8.4) | 0 (0) | 0 (0) | |
| Colon/metastatic (%) | 49 (65.3) | 74 (96.1) | 62 (92.5) | 145 (86.8) | 12 (100) | 67 (97.1) | |
| Other/unknown (%) | 14 (18.7) | 3 (3.9) | 0 (0) | 8 (4.8) | 0 (0) | 2 (2.9) | |
| ASA class | 0.007 | ||||||
| I/II (%) | 14 (18.7) | 12 (15.6) | 22 (32.8) | 47 (28.1) | 6 (50.0) | 26 (37.7) | |
| III/IV (%) | 61 (81.3) | 65 (84.4) | 45 (67.2) | 120 (71.9) | 6 (50.0) | 43 (62.3) | < 0.001 |
| Wound class | |||||||
| Clean (%) | 11 (14.7) | 0 (0) | 1 (1.5) | 17 (10.2) | 1 (8.3) | 3 (4.4) | |
| Clean-contaminated (%) | 61 (81.3) | 70 (90.9) | 65 (97.0) | 150 (89.8) | 11 (91.7) | 65 (94.2) | |
| Contaminated (%) | 3 (4.0) | 7 (9.1) | 1 (1.5) | 0 (0) | 0 (0) | 1 (1.5) | |
| Smoker (%) | 10 (13.3) | 15 (19.5) | 7 (10.5) | 17 (10.2) | 2 (16.7) | 4 (5.8) | 0.171 |
| Disseminated cancer (%)a | 48 (64.0) | 69 (89.6) | 45 (67.2) | 130 (77.8) | 10 (83.3) | 59 (85.5) | < 0.001 |
| Medical comorbidities | 0.077 | ||||||
| None (%) | 40 (53.3) | 39 (50.7) | 43 (64.2) | 84 (50.3) | 7 (58.3) | 49 (71.0) | |
| One (%) | 26 (34.7) | 32 (41.6) | 18 (26.9) | 58 (34.7) | 5 (41.7) | 13 (18.8) | |
| Two or more (%) | 9 (12.0) | 6 (7.8) | 6 (9.0) | 25 (15.0) | 0 (0) | 7 (10.1) |
HAIP: Hepatic artery infusion pump; SD: standard deviation; ASA: American Society of Anesthesiologists
Defined as a diagnosed cancer with spread to a major organ necessitating active treatment
Among patients who underwent HAIP placement, DSM occurred in 10.7% of those who did not receive a concurrent procedure. For patients who underwent concurrent procedures, rates of DSM were higher, as expected, for increasingly complex concurrent procedures: 19.2% [aOR (adjusted odds ratio) 2.20, 95% CI 0.91–5.29] for minor hepatic procedures, 22.1% (aOR 2.21, 95% CI 0.83–5.86) for colorectal resection, 23.2% (aOR 2.85, 95% CI 1.05–7.76) for concurrent minor hepatic plus colorectal procedures, 28.4% (aOR 3.79, 95% CI 1.44–9.99) for major hepatic resection, and 41.7% (aOR 6.77, 95% CI 1.59–28.85) for concurrent major hepatic plus colorectal resection. Rates of infectious complications, wound complications, LOS, and operative time also varied for the individual cohorts (Table 2).
TABLE 2.
Adjusted outcomes by concurrent surgery type compared with HAIP placement alone
| Death and serious morbidity | n (%) | Adjusted odds ratio (95% CI) |
P value |
|---|---|---|---|
| HAIP alone (n = 75) | 8 (10.7) | 1.00 (REF) | - |
| Minor liver (n = 167) | 32 (19.2) | 2.20 (0.91–5.29) | 0.080 |
| Colorectal (n = 77) | 17 (22.1) | 2.21 (0.83–5.86) | 0.111 |
| Colorectal plus minor liver (n = 69) | 16 (23.2) | 2.85 (1.05–7.76) | 0.040 |
| Major liver (n = 67) | 19 (28.4) | 3.79 (1.44–9.99) | 0.007 |
| Colorectal and major liver (n = 12) | 5 (41.7) | 6.77 (1.59–28.85) | 0.010 |
| Wound complications | n (%) | Adjusted odds ratio (95% CI) |
P value |
| HAIP alone (n = 75) | 2 (2.7) | 1.00 (REF) | - |
| Minor liver (n = 167) | 17 (10.2) | 4.39 (0.94–20.57) | 0.061 |
| Colorectal (n = 77) | 13 (17.9) | 6.13 (1.24–30.17) | 0.026 |
| Colorectal plus minor liver (n = 69) | 10 (14.5) | 6.86 (1.32–35.50) | 0.022 |
| Major liver (n = 67) | 10 (14.9) | 7.58 (1.49–38.59) | 0.015 |
| Colorectal and major liver (n = 12) | 4 (33.3) | 18.94 (2.68–134.00) | 0.003 |
| Infectious complications | n (%) | Adjusted odds ratio (95% CI) |
P value |
| HAIP alone (n = 75) | 3 (4.0) | 1.00 (REF) | - |
| Minor liver (n = 167) | 19 (11.4) | 3.32 (0.91–12.05) | 0.069 |
| Colorectal (n = 77) | 12 (15.6) | 4.07 (1.03–16.03) | 0.045 |
| Colorectal plus minor liver (n = 69) | 13 (18.8) | 6.66 (1.67–26.60) | 0.007 |
| Major liver (n = 67) | 11 (16.4) | 5.34 (1.35–21.14) | 0.017 |
| Colorectal and major liver (n = 12) | 4 (33.3) | 13.45 (2.34–77.26) | 0.004 |
| Length of stay (days) | Median (IQR) | Adjusted incidence rate ratio (95% CI) |
P value |
| HAIP alone (n = 75) | 5 (4–7) | 1.00 (REF) | - |
| Minor liver (n = 167) | 6 (5–7) | 1.29 (1.15–1.46) | < 0.001 |
| Colorectal (n = 77) | 7 (6–9) | 2.00 (1.76–2.28) | < 0.001 |
| Colorectal plus minor liver (n = 69) | 7 (6–10) | 1.81 (1.58–2.08) | < 0.001 |
| Major liver (n = 67) | 7 (6–10) | 1.86 (1.63–2.12) | < 0.001 |
| Colorectal and major liver (n = 12) | 10 (7–15) | 2.94 (2.44–3.54) | < 0.001 |
| Operative time (minutes) | Mean (SD) | Mean difference β (95% CI) |
P value |
| HAIP alone (n = 75) | 186.6 (58.9) | 0.0 (REF) | - |
| Minor liver (n = 167) | 252.4 (83.2) | 62.5 (37.3–87.7) | < 0.001 |
| Colorectal (n = 77) | 326.4 (120.8) | 130.5 (100.8–160.1) | < 0.001 |
| Colorectal plus minor liver (n = 69) | 336.2 (84.9) | 143.4 (112.5–174.4) | < 0.001 |
| Major liver (n = 67) | 307.7 (94.3) | 122.5 (92.0–152.9) | < 0.001 |
| Colorectal and major liver (n = 12) | 398.4 (94.8) | 209.4 (154.0–264.8) | < 0.001 |
HAIP: Hepatic artery infusion pump; CI: confidence interval; IQR: interquartile range; SD: standard deviation
In the propensity-matched analysis of the additive risk of HAIP placement, 388 patients who underwent HAIP placement with a concurrent hepatic and/or colorectal procedure were matched with 1161 patients who underwent hepatic and/or colorectal surgery without HAIP placement. After matching, the two cohorts did not differ with respect to any predictor (Table 3). No difference in DSM was seen for HAIP placement with concurrent hepatic and/or colorectal operations compared with hepatic and/or colorectal operations alone (22.7% vs 26.0%, OR 0.82, 95% CI 0.62–1.09). Additionally, no difference was seen in incidence of wound complications (13.7% vs 10.6%, OR 1.35, 95% CI 0.94–1.93) or infectious complications (15.0% vs 14.6%, OR 1.03, 95% CI 0.73–1.44) between these two groups (Table 4). LOS was increased among the patients who underwent HAIP placement with concurrent surgery compared with those who underwent similar surgeries without HAIP placement (median 6 vs 5 days, IRR 1.20, 95% CI 1.08–1.33), as was operative time (mean 294.7 vs 239.8 min, β = 54.9, 95% CI 42.8–67.0).
TABLE 3.
Characteristics of cohorts with and without HAIP placement following propensity score matching
| HAIP (n = 388) | No HAIP (n = 1161) | P value | |
|---|---|---|---|
| Mean age (SD) | 54.1 (11.2) | 54.6 (12.6) | 0.507 |
| Sex | 0.278 | ||
| Male (%) | 230 (59.3) | 725 (62.5) | |
| Female (%) | 158 (40.7) | 436 (37.6) | |
| Mean body mass index (SD) | 28.3 (5.9) | 28.0 (6.2) | 0.509 |
| Race | 0.960 | ||
| Non-Hispanic White (%) | 299 (77.1) | 893 (76.9) | |
| Non-Hispanic Black (%) | 17 (4.4) | 56 (4.8) | |
| Hispanic (%) | 14 (3.6) | 46 (4.0) | |
| Other/unknown (%) | 58 (15.0) | 166 (14.3) | |
| Cancer site | 0.961 | ||
| Primary liver malignancy (%) | 19 (4.9) | 57 (4.9) | |
| Colon/metastasis (%) | 359 (92.5) | 1077 (92.8) | |
| Other/unknown (%) | 10 (2.6) | 27 (2.3) | |
| ASA class | 0.290 | ||
| I/II (%) | 112 (28.9) | 303 (26.1) | |
| III/IV (%) | 276 (71.1) | 858 (73.9) | |
| Wound class | 0.727 | ||
| Clean (%) | 22 (5.7) | 62 (5.3) | |
| Clean-contaminated (%) | 357 (92.0) | 1079 (92.9) | |
| Contaminated (%) | 9 (2.3) | 20 (1.7) | |
| Smoker (%) | 45 (11.6) | 129 (11.1) | 0.781 |
| Disseminated cancer (%) | 309 (79.6) | 924 (79.6) | 1.000 |
| Medical comorbidities | 0.751 | ||
| None (%) | 218 (59.2) | 628 (54.1) | |
| One (%) | 126 (32.5) | 400 (34.5) | |
| Two or more (%) | 44 (11.3) | 133 (11.4) |
HAIP: Hepatic artery infusion pump; SD: standard deviation; ASA: American Society of Anesthesiologists
TABLE 4.
Outcomes for Patients with and without HAIP placement
| HAIP (n = 388) | No HAIP (n = 1161) |
Adjusted odds ratio (95% CI) |
|
|---|---|---|---|
| Death and serious morbidity (%) | 88 (22.7) | 302 (26.0) | 0.82 (0.62–1.09) |
| Wound complication (%) | 53 (13.7) | 123 (10.6) | 1.35 (0.94–1.93) |
| Infectious complications (%) | 58 (15.0) | 169 (14.6) | 1.03 (0.73–1.44) |
| Adjusted incidence rate ratio (95% CI) |
|||
| Median length of stay (IQR) | 6 (5–9) | 5 (4–7) | 1.20 (1.08–1.33) |
| Adjusted difference, β (95% CI) |
|||
| Mean operative time, minutes (SD) | 294.7 (101.3) | 239.8 (129.7) | 54.9 (42.8–67.0) |
HAIP: Hepatic artery infusion pump; CI: confidence interval; IQR: interquartile range; SD: standard deviation
DISCUSSION
The safety of concurrently performing HAIP placement with major surgery versus staging the procedures is an important consideration in the treatment of certain primary and secondary hepatic malignancies, and prior literature on the topic is limited. Although the risks of complications in this study vary for HAIP placement with different combinations of hepatic and colorectal operations, concurrent placement of a HAIP does not appear to increase the likelihood of complications. In other words, short-term operative risks appear to be primarily attributable to the other procedures performed, rather than to HAIP placement. The secondary measures of morbidity, including infectious and wound complications, demonstrate similar relationships among the procedure cohorts. The only outcomes significantly increased with the addition of HAIP placement are total hospital LOS and operative time, as expected.
To our knowledge, this is the first study to present data regarding complications rates when HAIP placement is performed with multiple categories of concurrent procedures. A prior study by Allen et al. reported a single institution experience with HAIP placement, which included an evaluation of surgical complication rates, LOS, and operative time for HAIP placement with concurrent colon resection.13 In addition to being limited to a single institution cohort, this study did not assess types of concurrent procedures aside from colon resection. Our study adds new and important data to the literature. It is important to understand expected rates of complications for concurrent HAIP placement with various hepatic and/or colorectal procedures. In the cohort with the highest morbidity in our study, consisting of patients who underwent concurrent major hepatic plus colorectal resection at the time of HAIP placement, the overall incidence of DSM was over 40% and median LOS was 10 days. Less extensive procedures had lower rates of complications and shorter hospital stays; however, significant morbidity was still observed in all cohorts. These results warrant consideration when evaluating different treatment strategies and counseling patients about the expected outcomes of combined procedures compared with alternative options. In general, treatment plans for advanced hepatic malignancy should be tailored to each individual patient.
In addition to demonstrating variation in risk with the different procedure combinations, this study evaluates whether synchronous HAIP placement at the time of concurrent procedures significantly contributes to the overall risk. By answering this question, this study provides new insights which may aid in decision making regarding operative approach and sequencing of concurrent procedures with HAIP placement. Although prior studies have shown that regional hepatic chemotherapy is beneficial for certain malignancies, the literature provides minimal guidance about the timing of HAIP placement in relation to other procedures indicated for oncologic treatment.26-31 Based on this study, although performing a major hepatic and/or colorectal surgery at the time of HAIP placement is associated with a higher risk of complications compared with HAIP placement alone, there was no increase in the morbidity when the combined procedure is compared with hepatic and/or colorectal resection alone. This suggests that, when both HAIP placement and additional procedures are indicated, the added morbidity of performing the procedures simultaneously is minimal compared with the sum of the individual operations. Therefore, decisions regarding the operative approach should primarily be determined based on the risk of the concurrent procedures. This is accompanied by the caveat that, as expected, LOS is increased when the procedures are performed concurrently, which may be attributable to postoperative radiologic testing of the newly implanted device or postoperative ileus due to longer operative time. However, when considering that staged operations would require two separate hospitalizations to recover from two major operations, the slightly longer stay for a combined procedure is likely of little consequence.
This study has several limitations. First, in this retrospective study, it is not possible to completely control for confounding factors. Therefore, this study can only demonstrate association, not causation. Second, the ACS NSQIP dataset does not permit identification of individual patients who have undergone more than one operation. This prevented the creation of a comparison group of patients who underwent staged HAIP placement with hepatic and/or colorectal resection. In place of such a comparison group, this study used propensity score matching to assess the morbidity of HAIP placement with concurrent hepatic and/or colorectal procedures through comparison with a surrogate group of patients who underwent the concurrent procedures alone. Comparison with this group in place of a cohort who underwent multiple operations in a staged fashion potentially introduces bias towards rejecting the null hypothesis. However, as no difference was seen in complication rates, this concern appears to be disproven. Finally, HAIP placement is performed at a small number of highly specialized centers and results may not be generalizable. This is especially important when considering new or small HAIP centers, as outcomes may differ substantially from those presented here. However, due to the complexity of building and instituting a HAIP program, it is anticipated that many cases will continue to come from experienced centers where the findings of this study are likely applicable.
CONCLUSION
We demonstrated that short-term outcomes of HAIP placement vary widely depending on the concurrent hepatic and/or colorectal resections that may be performed. Outcomes following HAIP placement are driven by the concurrent hepatic and/or colorectal procedure(s), rather than the HAIP. Decisions regarding HAIP placement should be considered primarily with respect to the risk of the concurrent operations performed, along with patient-and disease-specific factors.
ACKNOWLEDGMENTS
DJB is supported by the Veteran’s Administration (I01HX002290). ADY is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (K08HL145139). RPM is supported by the Agency for Healthcare Research and Quality (K12HS026385) and an Institutional Research Grant from the American Cancer Society (IRG-18-163-24). DM provides advising or speaking services for Amgen Inc., Bristol Meyers Squibb Company, Exelixis Inc., Eisai Co. Ltd., and Genentech Inc. and receives research funding from Merck & Co. Inc. and Oncolytics Biotech Inc.
Appendix
APPENDIX 1:
CURRENT PROCEDURE TERMINOLOGY (CPT) CODES INCLUDED IN EACH COHORT
| Operative cohort | CPT codes |
|---|---|
| Hepatic artery infusion pump placement | 36260 |
| Colon or rectum resection | 44120, 44140, 44141, 44143, 44144, 44145, 44146, 44147, 44150, 44151, 44155, 44156, 44157, 44158, 44160, 44204, 44205, 44206, 44207, 44208, 44210, 44211, 44212, 45110, 45111, 45112, 45113, 45114, 45116, 45119, 45120, 45121, 45123, 45126, 45130, 45135, 45160, 45395, 45397, 45402, 45550 |
| Major hepatic resection | 47122, 47125, 47130 |
| Minor hepatic procedures | 47100, 47120, 47370, 47371, 47379, 47380, 47381, 47382, 47383, 47399 |
Footnotes
DISCLOSURE DM provides advising or speaking services for Amgen Inc., Bristol Meyers Squibb Company, Exelixis Inc., Eisai Co. Ltd., and Genentech Inc. and receives research funding from Merck & Co. Inc. and Oncolytics Biotech Inc.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
REFERENCES
- 1.National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology. Colon Cancer (Version 1.2020) https://www.nccn.org/professionals/physician_gls/pdf/colon.pdf. Accessed February 11, 2020.
- 2.National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology. Hepatobiliary Cancers (Version 4.2019). https://www.nccn.org/professionals/physician_gls/pdf/hepatobiliary.pdf. Accessed February 11, 2020.
- 3.Adam R Chemotherapy and surgery: new perspectives on the treatment of unresectable liver metastases. Ann Oncol. 2003;14 Suppl 2:ii13–16. [DOI] [PubMed] [Google Scholar]
- 4.Khan SA, Davidson BR, Goldin RD, et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: an update. Gut. 2012;61(12):1657–1669. [DOI] [PubMed] [Google Scholar]
- 5.Datta J, Narayan RR, Kemeny NE, D’Angelica MI. Role of hepatic artery infusion chemotherapy in treatment of initially unresectable colorectal liver metastases: a review. JAMA Surg. 2019;154(8):768–776. [DOI] [PubMed] [Google Scholar]
- 6.Schlick CJR, Merkow RP, Bentrem DJ. Nonresectional regional therapies for metastatic colorectal cancer to the liver. J Surg Oncol. 2019;119(5):636–641. [DOI] [PubMed] [Google Scholar]
- 7.Karanicolas PJ, Metrakos P, Chan K, et al. Hepatic arterial infusion pump chemotherapy in the management of colorectal liver metastases: expert consensus statement. Curr Oncol. 2014;21(1):e129–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ko YJ, Karanicolas PJ. Hepatic arterial infusion pump chemotherapy for colorectal liver metastases: an old technology in a new era. Curr Oncol. 2014;21(1):e116–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ellis RJ, Angelos P, Jarnagin WR, Kemeny NE, Merkow RP. Abrupt Discontinuation of the Codman Hepatic Artery Infusion Pump: Considerations in the Era of Precision Medicine. J Am Coll Surg. 2019;229(2):217–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Liver-directed chemotherapy device is saving lives. The Globe and Mail. 2013. https://www.theglobeandmail.com/life/health-and-fitness/advsunnybrook/healing-the-future/liver-directed-chemotherapy-device-is-saving-lives/article12379031/. Accessed February 2020. [Google Scholar]
- 11.Buisman FE, Homs MYV, Grunhagen DJ, et al. Adjuvant hepatic arterial infusion pump chemotherapy and resection versus resection alone in patients with low-risk resectable colorectal liver metastases—the multicenter randomized controlled PUMP trial. BMC Cancer. 2019;19(1):327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Dizon DS, Schwartz J, Kemeny N. Regional chemotherapy: a focus on hepatic artery infusion for colorectal cancer liver metastases. Surg Oncol Clin N Am. 2008;17(4):759–771, viii. [DOI] [PubMed] [Google Scholar]
- 13.Allen PJ, Nissan A, Picon AI, et al. Technical complications and durability of hepatic artery infusion pumps for unresectable colorectal liver metastases: an institutional experience of 544 consecutive cases. J Am Coll Surg. 2005;201(1):57–65. [DOI] [PubMed] [Google Scholar]
- 14.Daly JM, Kemeny N, Oderman P, Botet J. Long-term hepatic arterial infusion chemotherapy. Anatomic considerations, operative technique, and treatment morbidity. Arch Surg. 1984;119(8):936–941. [DOI] [PubMed] [Google Scholar]
- 15.Martin RC, Edwards MJ, McMasters KM. Morbidity of adjuvant hepatic arterial infusion pump chemotherapy in the management of colorectal cancer metastatic to the liver. Am J Surg. 2004;188(6):714–721. [DOI] [PubMed] [Google Scholar]
- 16.Khuri SF, Daley J, Henderson W, et al. The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA Surgical Quality Improvement Program. Ann Surg. 1998;228(4):491–507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Khuri SF, Henderson WG, Daley J, et al. Successful implementation of the Department of Veterans Affairs’ National Surgical Quality Improvement Program in the private sector: the Patient Safety in Surgery study. Ann Surg. 2008;248(2):329–336. [DOI] [PubMed] [Google Scholar]
- 18.Khuri SF, Henderson WG, Daley J, et al. The patient safety in surgery study: background, study design, and patient populations. J Am Coll Surg. 2007;204(6):1089–1102. [DOI] [PubMed] [Google Scholar]
- 19.Cohen ME, Ko CY, Bilimoria KY, et al. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg. 2013;217(2):336–346 e331. [DOI] [PubMed] [Google Scholar]
- 20.Ingraham AM, Richards KE, Hall BL, Ko CY. Quality improvement in surgery: the American College of Surgeons National Surgical Quality Improvement Program approach. Adv Surg. 2010;44:251–267. [DOI] [PubMed] [Google Scholar]
- 21.National Quality Forum. NQF #0697: Risk Adjusted Case Mix Adjusted Elderly Surgery Outcomes Measure. http://www.qualityforum.org. Published 2017. Accessed 4 May 2020.
- 22.National Quality Forum. NQF #0706: Risk Adjusted Colon Surgery Outcome Measure. http://www.qualityforum.org. Published 2017. Accessed 4 May 2020.
- 23.Merkow RP, Hall BL, Cohen ME, et al. Validity and feasibility of the American College of Surgeons colectomy composite outcome quality measure. Ann Surg. 2013;257(3):483–489. [DOI] [PubMed] [Google Scholar]
- 24.Bilimoria KY, Chung JW, Hedges LV, et al. National Cluster-Randomized Trial of Duty-Hour Flexibility in Surgical Training. N Engl J Med. 2016;374(8):713–727. [DOI] [PubMed] [Google Scholar]
- 25.SAS Institute Inc. SAS/STAT(R) 14.2 User’s Guide. In: Cary, NC: SAS Institute Inc; 2016. [Google Scholar]
- 26.D’Angelica MI, Correa-Gallego C, Paty PB, et al. Phase II trial of hepatic artery infusional and systemic chemotherapy for patients with unresectable hepatic metastases from colorectal cancer: conversion to resection and long-term outcomes. Ann Surg. 2015;261(2):353–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Konstantinidis IT, Groot Koerkamp B, Do RK, et al. Unresectable intrahepatic cholangiocarcinoma: systemic plus hepatic arterial infusion chemotherapy is associated with longer survival in comparison with systemic chemotherapy alone. Cancer. 2016;122(5):758–765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lim A, Le Sourd S, Senellart H, et al. Hepatic Arterial Infusion Chemotherapy for Unresectable Liver Metastases of Colorectal Cancer: A Multicenter Retrospective Study. Clin Colorectal Cancer. 2017;16(4):308–315. [DOI] [PubMed] [Google Scholar]
- 29.Lorenz M, Muller HH, Schramm H, et al. Randomized trial of surgery versus surgery followed by adjuvant hepatic arterial infusion with 5-fluorouracil and folinic acid for liver metastases of colorectal cancer. German Cooperative on Liver Metastases (Arbeitsgruppe Lebermetastasen). Ann Surg. 1998;228(6):756–762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Alberts SR, Roh MS, Mahoney MR, et al. Alternating systemic and hepatic artery infusion therapy for resected liver metastases from colorectal cancer: a North Central Cancer Treatment Group (NCCTG)/National Surgical Adjuvant Breast and Bowel Project (NSABP) phase II intergroup trial, N9945/CI-66. J Clin Oncol. 2010;28(5):853–858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Levi FA, Boige V, Hebbar M, et al. Conversion to resection of liver metastases from colorectal cancer with hepatic artery infusion of combined chemotherapy and systemic cetuximab in multicenter trial OPTILIV. Ann Oncol. 2016;27(2):267–274. [DOI] [PubMed] [Google Scholar]