Postmastectomy breast reconstruction options include a variety of prosthetic and autologous techniques.1–9 Although alloplastic reconstruction remains most common, microvascular autologous-based reconstruction has demonstrated good aesthetic results with high long-term patient satisfaction in a single-stage procedure.5–12 While possessing numerous benefits, microsurgical breast reconstruction is associated with longer operative times and postoperative length of inpatient hospital stay compared with implant-based reconstruction.11,12
Meanwhile, length of postoperative hospitalization is becoming an increasingly utilized surgical outcome metric.13,14 With regard to autologous breast reconstruction, increased length of hospital stay has unsurprisingly been linked to increasing health-care cost.15 Multiple studies have also evaluated the influence of specific interventions or enhanced recovery pathways on outcomes after autologous breast reconstruction, including length of hospital stay.16–18
Despite the focus on maximizing patient outcomes while minimizing hospital-based cost, there is a lack of evidence examining risk factors for increasing length of postoperative hospital stay after autologous breast reconstruction. Such information would benefit hospital systems and individual surgeons alike by identifying actionable factors that can be addressed to improve outcomes and refine risk stratification of patients presenting to discuss surgical options for postmastectomy breast reconstruction. We therefore seek to elucidate factors associated with increased length of stay (LOS) after autologous breast reconstruction at a large, metropolitan center utilizing a case-control study design.
All patients undergoing autologous microsurgical breast reconstruction utilizing an abdominal donor site at NYU Langone Medical Center with 6 primary plastic surgeons over a 2-year time period from 2015 to 2017 were identified. Length of inpatient hospital stay was calculated for all patients. By definition, the day of surgery was denoted as postoperative day 0; postoperative day 1 started on the morning following the day of surgery. Patients were distributed into 2 groups: those with a LOS greater than or equal to 5 days (120 h) and those with a LOS of 4 days (96 h) or less. Patient demographics, intraoperative variables, and overall and individual reconstructive outcomes were collected, analyzed, and compared for the 2 groups in a case-control study design.19 Risk factors associated with longer LOS were thus identified.
With regard to reconstructive complications, individual complications were divided into 3 main categories: abdominal donor-site complications, breast-related complications, and microsurgical complications. Delayed wound healing of the abdominal donor site was defined by any wound breakdown that was treated either with local wound care or debridement, whether in an office or operative setting. Cellulitis was considered any superficial infection of the breasts or abdomen that was treated with either oral or intravenous (IV) antibiotics. Major mastectomy flap necrosis was defined as that managed with operative or office debridement, whereas minor mastectomy flap necrosis was that managed with local wound care.
Descriptive statistics and measures of central tendency were used to describe absolute and mean results, respectively. Student’s t tests were used to analyze binary data sets, whereas Chi-squared analysis was used to compare proportional responses. All statistical analysis was performed using GraphPad Software, Inc. (La Jolla, Calif.). P values of less than 0.05 were deemed significant.
All patients were transferred to a surgical step-down unit for flap checks after initial monitoring in the postanesthesia care unit. Flap checks were performed by surface Doppler signals and clinical assessment every 1 hour for the first 24 hours, and subsequently every 2 hours and 4 hours on postoperative days 2 and 3, respectively. Anticoagulation was started the evening of surgery with aspirin 300 mg per rectum or 325 mg orally followed by 81 mg orally for 1 month. Patients were also placed on 40 mg of low weight molecular heparin subcutaneously daily for venous thromboembolism prophylaxis while in-house.
On postoperative day 1, patients were given a diet and encouraged to get out of bed to a chair. Foley catheters were subsequently removed once patients were able to sit in a chair. IV fluids were stopped when patients were able to demonstrate adequate oral intake. On postoperative day 2, patients were encouraged to ambulate with physical therapy and were transferred to a regular room if flap checks, vital signs, and their overall clinical condition were stable. Subsequent progress and discharge were based on the patient’s ability to independently ambulate, achieve adequate pain control with oral pain medications, and tolerate a regular diet, and their confidence in their ability to function at home with whatever assistance was available. All aspects of postoperative management and discharge parameters are extensively discussed with patients preoperatively.
Patients generally received as needed oral narcotic pain medications, which was supplemented with standing or as needed acetaminophen, standing ketorolac and/or ibuprofen if there was no concern for excess bleeding, and IV narcotic medication for breakthrough pain. In the earlier study period, patients were placed on patient-controlled analgesia pumps that were discontinued the first postoperative morning. All patients were also given 5 mg of Valium orally scheduled every 8 hours for muscle spasm relief. Patients received low-dose aspirin as a scheduled medication for its antiplatelet rather than analgesic effect.
A total of 116 patients undergoing immediate or delayed abdominally based microsurgical breast reconstruction were identified. Of these, 86 (74.1%) had a LOS of 4 days (96 h) or less (mean: 3.70 d; range: 2–4 d), whereas 30 (25.9%) had a LOS of 5 days (120 h) or greater (mean: 5.50 d; range: 5–8 d). There was a significant difference in LOS between the groups (P < 0.0001). Overall LOS for the entire cohort of patients was 4.16 days (range: 2–8) (Fig. 1).
With regard to patient demographics and intraoperative factors, patients with a LOS of 5 days or greater were significantly more likely to have diabetes mellitus (P < 0.0001), undergo bilateral reconstruction (P = 0.0003) and total mastectomy (P < 0.0001), and have a longer operative time (P < 0.0001). These patients were significantly less likely to undergo postoperative radiation (P = 0.0421). There were no significant differences between these 2 groups in terms of age (P = 0.0714), body mass index (P = 0.4541), smoking history (P = 0.6829), timing of reconstruction (P = 0.4166), double-attending cases (P = 0.8329), or transversus abdominis plane (TAP) block (P = 0.2329). Notably, there was no significant difference between the groups in terms of follow-up time or time since breast reconstruction (P = 0.0600) (Table 1).
With regard to reconstructive complications, there were equivalent rates of overall complications between patients with a LOS of 4 days or less and those with a LOS of 5 days or greater (47.7% vs 53.3%; P = 0.5367). Patients with LOS of 5 days of greater were significantly more likely to experience abdominal donor-site abscess (P < 0.0001), breast hematoma (P = 0.0186), and return to the operating room for flap compromise (P < 0.0001). Notably, there were no instances of complete flap loss as all cases in which a return to the operating room was required resulted in flap salvage. Inherently, there was no significant difference between the groups in this regard (P = 1.000) (Table 2). No patients returned to the operating room for any reason other than flap compromise, such as for debridement of mastectomy flap necrosis.
Microsurgical autologous breast reconstruction has numerous advantages over implant-based techniques, which remain the most commonly employed.5 These advantages include single-stage reconstruction, good aesthetic results, especially in unilateral reconstructions, and improved patient satisfaction compared with alloplastic reconstruction.5–12,20 Despite these benefits, autologous reconstruction is also associated with longer operative times and postoperative length of inpatient hospital stay.11,12 Longer postoperative hospital stays with autologous breast reconstruction have further been linked to greater initial health-care cost.11,12,15
In the current health-care climate, multiple metrics have been evaluated in an effort to improve patient outcomes in a most cost-efficient manner.13,14 Postsurgical length of inpatient hospital stay has become a commonly utilized metric in this regard.13,14,16–18 The rise in implementation of enhanced recovery pathways after autologous breast reconstruction has mirrored the effort to reduce postoperative LOS and associated costs.13,14,16–19,21 For instance, Afonso et al. demonstrated significantly reduced length of hospital stay after autologous breast reconstruction with implementation of an enhanced recovery pathway comprised of reducing opioid-centered pain management, regional abdominal blocks, and goal-directed fluid therapy, among other changes.17
Although these studies have rightfully examined methods to reduce hospital LOS after autologous breast reconstruction, system- and patient-specific risk factors for increased LOS remain to be elucidated. We therefore sought to identify such risk factors in patients undergoing autologous breast reconstruction using an abdominal donor site, which remains the most frequently utilized.22,23 Further, all cases were performed at a large, metropolitan hospital center with a high microsurgical volume. Therefore, surgeon and care team familiarity and experience in managing patients undergoing microsurgical breast reconstruction were constant and expectantly high throughout the study period.
Postoperative length of inpatient hospital stay for the overall cohort of 116 patients was 4.16 days. This compares very favorably with the literature as it is nearly equivalent to the LOS demonstrated after implementation of an enhanced recovery pathway for abdominally based autologous breast reconstruction in 1 recent study from a large cancer center.17 Given this natural inflection point, the groups were then divided into those with a LOS ≤4 days (96 h) and greater than 4 days (≥5 d; 120 h). This case-control study design is advantageous as it is particularly well suited to illuminate various etiologic factors contributing to longer LOS.19
Diabetes mellitus was identified as a patient-specific factor increasing length of hospital stay after autologous breast reconstruction. This can be expected as diabetes increases overall risk of wound healing issues, may complicate postoperative medical care, and may be associated with additional comorbidities that can complicate care and delay patient discharge. Wound complications due to diabetes mellitus tend to be subacute or chronic issues. However, patients with diabetes have been shown to be at greater risk for acute wound complications as well.24 Although these may not lead to a need for immediate reoperation, they can delay patient readiness for discharge. Interestingly, patients with longer postoperative LOS were also identified to have significantly less postoperative radiation. The etiology of this finding is unclear and observational, although it is noted that there was a nonsignificant trend toward more prophylactic mastectomies in the shorter LOS group. Prior radiation was not identified as a risk factor for greater LOS. Importantly, body mass index, which has been identified as a risk factor for increased pain after breast surgery, did not affect postoperative LOS.25 This may be due to the finding that the average patient in both groups was classified as overweight according to body mass index (>25 kg/m2). The independent effect of body mass index on LOS after microsurgical breast reconstruction remains to be defined.
In examining operative factors for increasing postoperative LOS, bilateral reconstruction, total mastectomy, and increased operative time were identified as risk factors. All of these factors implicate more technically challenging and involved procedures, which may be associated with greater postoperative pain and complexity of care, thus leading to longer postoperative stays.26 Notable factors that were not found to increase LOS included timing of reconstruction (immediate vs delayed) and amount of muscle taken with the abdominal flap. Taken overall, the oncologic portion of an immediate breast reconstruction procedure seems to influence LOS less than the reconstructive portion, as may be expected. Further investigation may be required to further delineate the unique attributes of patient subsets undergoing immediate or delayed breast reconstruction as related to postoperative outcomes. However, given the results discussed earlier, the type of mastectomy performed in patients undergoing immediate breast reconstruction does seem to influence the LOS metric.
Interestingly, the utilization of regional TAP blocks was not found to be more likely in the group with shorter LOS. As a component of enhanced recovery pathways, TAP blocks have been shown to help reduce LOS.17 However, the independent effects of regional blocks on postoperative metrics after abdominally based autologous breast reconstruction remain to be completely evaluated. Given the beneficial effects of TAP blocks in reducing acute postoperative pain, we have continued to utilize ultrasound guided TAP blocks in this patient population.21
Overall complications were not found to influence LOS. Only abdominal donor-site abscess, breast hematoma (both operative and nonoperative), and return to operating room for flap compromise were identified as risk factors in patients with increased LOS. Hematoma and return to operating room are acute complications, which can be expected to portend a more complicated hospital stay. Abdominal donor-site abscess is a subacute complication that would not necessarily be expected to increased immediate postoperative LOS. However, this complication occurred in a single patient in the cohort with a greater LOS. Additionally, the same patient also experienced a breast hematoma and subsequent return to the operating room for impending venous compromise, likely contributing to this correlation.
Lastly, some patients may elect to stay an additional hospital day despite otherwise meeting discharge criteria. Preoperative counseling is critical to set objective discharge parameters so that patients may be aware of when discharge may be expected and to mitigate these medically unnecessary additional hospital days. Unfortunately, these factors are difficult to measure. However, the influence of preoperative patient counseling and expectation setting cannot be overstated as surgeons work to minimize hospital cost and LOS while simultaneously improving patient outcomes.
Wound healing complications and infections of the abdominal donor site and breast did not influence immediate postoperative LOS. It is notable that there were no flap failures and loss of reconstructions in the overall patient cohort. Five patients, all in the group with a longer LOS, required return to the operating room for flap compromise. However, all flaps were salvaged. Further, despite this deferential in flap compromise, rates of fat necrosis were low (≤5.5%) and equivalent between the 2 groups.
The findings of this study may be utilized on patient-, surgeon-, and hospital-system levels. With this information, plastic surgeons can more accurately counsel patients preoperatively both in regard to their expected outcomes and to emphasize expected discharge parameters. Postoperative metrics, including LOS, may thus further be optimized. Surgeons may also be better able to direct patients toward the most appropriate breast reconstruction modality in accordance with each patient’s individualized preferences. At a hospital-system level, these findings may better inform administrators with regard to expected postoperative recovery periods for patients undergoing autologous breast reconstruction. As higher risk patients are identified, specific resources may be allocated toward enhancing and optimizing their recovery in a cost-effective manner, especially in cases of bundled reimbursement.
Limitations of this study include its retrospective nature and limited sample size, and that outcomes, such as mastectomy flap necrosis, were defined by intervention and thus influenced by surgeon preference in management. Follow-up time was approximately 1 year or greater in each group. This is sufficient follow-up to identify acute and subacute outcomes expected to influence immediate postoperative LOS. Moreover, follow-up time was equivalent between the 2 groups, confirming that the findings herein are not a result of greater capture in 1 group. Multi-institutional collaboration to compare factors related to LOS after autologous breast reconstruction as a future direction of research would strengthen the findings discussed herein. Lastly, this case-control study design may associate factors with increased LOS; however, future research is warranted to further elucidate and delineate these associations.
In conclusion, multiple patient-specific, intraoperative, and postoperative factors are associated with increased LOS in immediate and delayed microsurgical breast reconstruction. With this information, expected norms in postoperative outcome metrics and guidelines to modify pre-, intra-, and postoperative protocols in patients undergoing autologous breast reconstruction can be established. Further, risk stratification and establishment of patient expectations by surgeons and patient care teams may be improved. The findings herein may therefore benefit hospital systems, plastic surgeons, and patients alike.
1. Choi M, Frey JD, Alperovich M, et al. “Breast in a day”: examining single-stage immediate, permanent implant reconstruction in nipple-sparing mastectomy. Plast Reconstr Surg. 2016;138:184e–191e.
2. Peled AW, Foster RD, Garwood ER, et al. The effects of acellular dermal matrix in expander-implant breast reconstruction after total skin-sparing mastectomy: results of a prospective practice improvement study. Plast Reconstr Surg. 2012;129:901e–908e.
3. Colwell AS, Tessler O, Lin AM, et al. Breast reconstruction following nipple-sparing mastectomy: predictors of complications, reconstruction outcomes, and 5-year trends. Plast Reconstr Surg. 2014;133:496–506.
4. Gfrerer L, Mattos D, Mastroianni M, et al. Assessment of patient factors, surgeons, and surgeon teams in immediate implant-based breast reconstruction outcomes. Plast Reconstr Surg. 2015;135:245e–252e.
5. Fischer JP, Nelson JA, Au A, et al. Complications and morbidity following breast reconstruction–a review of 16,063 cases from the 2005-2010 NSQIP datasets. J Plast Surg Hand Surg. 2014;48:104–114.
6. Fujimoto H, Ishikawa T, Satake T, et al. Donor site selection and clinical outcomes of nipple-areola skin-sparing mastectomy with immediate autologous free flap reconstruction: a single-institution experience. Eur J Surg Oncol. 2016;42:369–375.
7. Levine SM, Snider C, Gerald G, et al. Buried flap reconstruction after nipple-sparing mastectomy: advancing toward single-stage breast reconstruction. Plast Reconstr Surg. 2013;132:489e–497e.
8. Tanna N, Broer PN, Weichman KE, et al. Microsurgical breast reconstruction for nipple-sparing mastectomy. Plast Reconstr Surg. 2013;131:139e–147e.
9. Yang SJ, Eom JS, Lee TJ, et al. Recipient vessel selection in immediate breast reconstruction with free abdominal tissue transfer after nipple-sparing mastectomy. Arch Plast Surg. 2012;39:216–221.
10. Becker H, Lind JG II, Hopkins EG. Immediate implant-based prepectoral breast reconstruction using a vertical incision. Plast Reconstr Surg Glob Open 2015;3:e412.
11. Sgarzani R, Negosanti L, Morselli PG, et al. Patient satisfaction and quality of life in DIEAP flap versus implant breast reconstruction. Surg Res Pract. 2015;2015:405163.
12. Fischer JP, Nelson JA, Cleveland E, et al. Breast reconstruction modality outcome study: a comparison of expander/implants and free flaps in select patients. Plast Reconstr Surg. 2013;131:928–934.
13. Robertson FC, Logsdon JL, Dasenbrock HH, et al. Transitional care services: a quality and safety process improvement program in neurosurgery. J Neurosurg. 2017:1–8.
14. Porter ME. What is value in health care? N Engl J Med. 2010;363:2477–2481.
15. Billig JI, Lu Y, Momoh AO, et al. A nationwide analysis of cost variation for autologous free flap breast reconstruction. JAMA Surg. 2017;152:1039–1047.
16. Bonde C, Khorasani H, Eriksen K, et al. Introducing the fast track surgery principles can reduce length of stay after autologous breast reconstruction using free flaps: a case control study. J Plast Surg Hand Surg. 2015;49:367–371.
17. Afonso A, Oskar S, Tan KS, et al. Is enhanced recovery the new standard of care in microsurgical breast reconstruction? Plast Reconstr Surg. 2017;139:1053–1061.
18. Batdorf NJ, Lemaine V, Lovely JK, et al. Enhanced recovery after surgery in microvascular breast reconstruction. J Plast Reconstr Aesthet Surg. 2015;68:395–402.
19. Hatchell AC, Farrokhyar F, Choi M. The misconception of case-control studies in the plastic surgery literature: a literature audit. Plast Reconstr Surg. 2017;139:1356e–1363e.
20. Eltahir Y, Werners LL, Dreise MM, et al. Which breast is the best? Successful autologous or alloplastic breast reconstruction: patient-reported quality-of-life outcomes. Plast Reconstr Surg. 2015;135:43–50.
21. Jablonka EM, Lamelas AM, Kim JN, et al. Transversus abdominis plane blocks with single-dose liposomal bupivacaine in conjunction with a nonnarcotic pain regimen help reduce length of stay following abdominally based microsurgical breast reconstruction. Plast Reconstr Surg. 2017;140:240–251.
22. Patel NG, Ramakrishnan V. Microsurgical tissue transfer in breast reconstruction. Clin Plast Surg. 2017;44:345–359.
23. Frey JD, Choi M, Salibian AA, et al. Comparison of outcomes with tissue expander, immediate implant, and autologous breast reconstruction in greater than 1000 nipple-sparing mastectomies. Plast Reconstr Surg. 2017;139:1300–1310.
24. Cunningham GR, Daoud D, Baimbridge S, et al. Effects of glycemia on immediate complications following CABG. Endocr Pract. 2013;19:928–936.
25. van Helmond N, Timmerman H, van Dasselaar NT, et al. High body mass index is a potential risk factor for persistent postoperative pain after breast cancer treatment. Pain Physician 2017;20:E661–E671.
26. Kulkarni AR, Pusic AL, Hamill JB, et al. Factors associated with acute postoperative pain following breast reconstruction. JPRAS Open 2017;11:1–13.