Factors Associated with Acute Postoperative Pain Following Breast Reconstruction

Summary

Introduction

Post-mastectomy breast reconstruction has become an increasingly important component of breast cancer treatment. Unfortunately, some patients experience severe postoperative pain, placing them at risk for increased clinical morbidity and the development of disabling chronic pain. In an attempt to identify at-risk patients, we prospectively evaluated patient characteristics and medical/surgical variables associated with more severe acute post-reconstruction pain.

Methods

Women (N = 2207; one-week 82.8% response rate) undergoing breast reconstruction were assessed for pain experience, anxiety, depression, and sociodemographic characteristics prior to surgery. Pain assessments were made preoperatively and postoperative at 1-week using validated survey instruments including the McGill Pain Questionnaire-Short Form (MPQ-SF), Numerical Pain Rating Scale (NPRS), and BREAST-Q Chest and Upper Body scale. Depressive symptoms and anxiety severity were assessed by the Patient Health Questionnaire and Generalized Anxiety Disorders Scale, respectively. Mixed-effects regression modeling was used to examine the relationships between patient characteristics and medical/surgical factors and 1-week postoperative pain.

Results

Younger age, bilateral reconstruction, and severity of preoperative pain, anxiety and depression were all associated with more severe acute postoperative pain on all the pain measures and BREAST-Q. Comparison of surgical procedure type indicated less severe postoperative pain for PTRAM, DIEP and SIEA reconstructive surgery compared to tissue expander/implant reconstruction.

Conclusions

This study identified patients at risk for greater acute postoperative pain following breast reconstruction. These findings will allow plastic surgeons to better tailor postoperative care to improve patient comfort, reduce clinical morbidity, and further enhance patient satisfaction with their surgical outcome.

Introduction

Breast reconstruction is a valuable adjunct to the treatment of breast cancer. ¹ For the majority of patients, acute postoperative pain following reconstruction is an expected surgical outcome that readily resolves with time and acute pain intervention. ² However, patients who suffer poorly controlled postoperative pain understandably report less satisfaction with their surgical experience. ^3,4 In addition, undermanaged acute postsurgical pain can lead to greater postoperative clinical morbidity and higher hospital costs.^5–8 For a surprising and unfortunate number of patients, acute postoperative pain persists beyond the normal course of postsurgical recovery, and more severe postoperative pain is established as a robust risk factor for the development of potentially debilitating chronic pain. ⁸ This is a concern for the cohort of women receiving breast reconstruction as one-fourth to one-half of these women report long-term persistent pain beyond the expected course of surgical recovery. ^9–12

Studies evaluating postoperative pain following breast reconstruction have generally examined the development of chronic rather than the severity of acute postsurgical pain. ^11–14 Indeed, clinical reviews and outcome studies describing complications following breast reconstruction uniformly fail to either consider or adequately address pain as an outcome variable,^15,16 even when assessing patient satisfaction with reconstruction surgery. ¹⁷ Recently, a number of investigations have begun to evaluate factors associated with severity of acute postoperative pain after reconstructive surgery. In an early descriptive study, Kroll et al. (2001) ¹⁸ reported that women undergoing free transverse rectus abdominis musculocutaneous flap reconstruction, compared to those choosing deep inferior epigastric perforator flap reconstruction, required higher doses of postoperative morphine. In a retrospective cohort study, Gassman and colleagues (2015) ¹⁹ evaluated immediate postoperative pain and opioid use in 378 women who had undergone tissue implant-based breast reconstruction. No significant difference in pain was observed comparing immediate and delayed postmastectomy reconstruction. Tissue expanders with larger volumes and initial fill were associated with higher narcotic use, while patients undergoing TE-based implant reconstruction required higher analgesic dosing than those with single-stage II-based reconstruction. In a separate report using a similar retrospective design with an enlarged data base, this same group ²⁰ compared acute postoperative pain-related parameters for women receiving either implant-based or flap-based reconstruction. The authors observed that both pain and analgesic use was greater in the implant-based group compared with autologous tissue–based reconstruction and for patients receiving bilateral rather than unilateral procedures. These studies provide preliminary evidence for variability of acute pain following reconstructive surgery based on procedure type and characteristics of surgical technique. However, to date no studies have examined variables that may predict more severe acute postoperative pain following breast reconstruction.

Drawing from a cross-section of studies from major surgery, an array of medical, sociodemographic and psychological variables have demonstrated significant relationship with severity of acute postoperative pain. Not surprisingly, type of surgery, particularly with surgical disruption of neurological pathways, has held consistent relationship with more severe acute postoperative pain. ^2,8,20,22 Younger age and female gender have similarly been found to predict severity of acute postsurgical pain. ^8,22,23 Preoperative pain experience, both pain sensitivity and patient self-report of pain intensity, also appear to be positively associated with acute postoperative pain. ^21,23,24 Among psychological factors, preoperative levels of depression, anxiety, and psychological vulnerability to aberrant pain perception (e.g., pain catastrophizing) are each significantly associated with greater postoperative pain intensity ^{21, 25–27} and are deemed risk factors for more severe acute postsurgical pain. ²⁸

Identifying those patients who are at higher risk for more severe postsurgical pain can galvanize efforts to optimize postoperative pain control for patients undergoing major surgery, ^8,29 including the development of an individualized presurgical protocol to aggressively manage postoperative pain with adequate analgesia and non-opioid pain interventions. ³⁰ To this end, this study sought to prospectively evaluate patients undergoing breast reconstruction to identify patient-specific and medical/surgical factors that predict higher levels of acute postoperative pain.

Methods

Study Setting and Participants

As part of the Mastectomy Reconstruction Outcomes Consortium (MROC) study, a prospective cohort of breast reconstruction patients was recruited at 11 centers in the United States and Canada. Fifty-eight plastic surgeons contributed patients to the study from Michigan, Ohio, Texas, New York, Massachusetts, Illinois, Washington DC, Manitoba, British Columbia, and Georgia. Women were eligible for study participation if they were over 18 years of age undergoing first-time unilateral or bilateral post-mastectomy breast reconstruction using one of eight common techniques: tissue expander/implant (TE/Implant), latissimus dorsi (LD), pedicle transverse rectus abdominis musculocutaneous flap (PTRAM), free TRAM flap including muscle-sparing (FTRAM), deep inferior epigastric perforator flap (DIEP), superficial inferior epigastric perforator flap (SIEA), superior gluteal artery perforator flap (SGAP), and inferior gluteal artery perforator flap (IGAP). Patients undergoing SGAP, and IGAP procedures were excluded from this analysis due to low numbers. Appropriate approval from the Institutional Review Board (IRB) at each participating site was obtained.

Study Design and Data Collection

Data were prospectively collected from medical records and patient surveys. Preoperatively, patients completed a series of sociodemographic and Patient-Reported Outcomes (PRO) questionnaires to assess patient characteristics, psychological status (depression and anxiety symptom severity), and pain. One week postoperatively, patients completed questionnaires to assess levels of depressive symptoms, anxiety symptoms and pain experience.

Measures

Depression

To assess severity of depressive symptoms, each subject completed the Patient Health Questionnaire (PHQ-9),³¹ a self-administered and frequently utilized inventory that assesses severity of depression in a non-psychiatric population. A PHQ-9 score ≥10 has a sensitivity of 88% and a specificity of 88% for major depression.³¹ A higher score indicates more severe symptom complaints.

Anxiety

Severity of anxiety symptoms was assessed by the Generalized Anxiety Disorders Scale (GAD-7),³² a brief self-report questionnaire also used as a validated screening tool for a non-psychiatric population, that identifies probable cases of generalized anxiety disorder. Higher scores denote more severe anxiety. A cut point score ≥10 demonstrates a sensitivity of 89% and specificity of 82% for a diagnosis of generalized anxiety disorder.²⁶ Although generalized anxiety and depression symptoms frequently co-occur, factor analysis of the PHQ-9 and GAD-7 indicates they measure distinct dimensions of mood disturbance.³²

Global Pain Ratings

Preoperative and postoperative pain intensity was assessed by the Numerical Pain Rating Scale (NPRS) ³³ which provides a single measure of overall pain intensity drawn from an ordinal numerical scale reflecting increasing pain severity, ranging from 0 to 10. A more broad measure of preoperative and postoperative pain experience was obtained by the McGill Pain Questionnaire-Short Form (MPQ-SF).³⁴ The MPQ-SF contains 15 descriptors of pain experience and provides a valid measure of the sensory and affective components of pain experience. The sensory pain rating (MPQ-Sensory Rating) quantifies the sensory dimensions of pain experience with a range of scores from 0 to 33 while the affective pain rating (MPQ-Affective Rating) provides a measure of the subjective unpleasantness or suffering associated with pain and can range from 0 to 12. For both the MPQ-SF and NPRS, higher scores indicate more severe pain experience.

Chest and Upper Body Pain and Dysfunction

In order to evaluate specific symptoms of pain and discomfort associated with reconstructive surgery, patients also completed The BREAST-Q Physical Well-being: Chest and Upper Body Scale, ^35,36 a validated condition-specific instrument that captures chest and upper body symptoms among women following breast reconstruction, including pain or tightness in the breast area and difficulty with mobility and physical activity. Higher scores on the BREAST-Q indicate better physical well-being and less pain.

Medical and Surgical Procedural Variables

To determine if postoperative pain is related to selected medical and surgical factors associated with reconstructive surgery, patients were identified by their body mass index (BMI) and lymph node status (none, sentinel lymph node biopsy, axillary lymph node dissection) at time of reconstruction. Patients were further categorized in terms of the following surgical procedure characteristics: laterality (unilateral vs. bilateral), timing of mastectomy to reconstruction (immediate vs. delayed), and mastectomy type (prophylactic vs. therapeutic).

Statistics

Mixed-effects regression modeling was used to assess the relationship between patient-specific factors (independent variables) and 1-week postoperative pain (dependent variable). Four separate models were fit for 1-week postoperative pain or pain-related outcomes as assessed by each of the following measures: MPQ-SF sensory component, MPQ-SF affective component, NPRS and BREAST-Q Chest and Upper Body scale. All models included the preoperative values of the dependent variable to adjust for baseline differences in pain level. The independent variables included age, race, preoperative anxiety (GAD-7), and preoperative depression (PHQ-9). The independent variables also included a number of surgical procedure factors including laterality (unilateral vs. bilateral), timing (immediate vs. delayed), mastectomy type (prophylactic vs. therapeutic), procedure type (TE/Implant, LD, PTRAM, Free/MSTRAM, DIEP, SIEA), and lymph node status (none (26%), sentinel lymph node biopsy (46%), and axillary lymph node dissection (27%) at the time of reconstruction. All models also included random intercepts for centers (hospitals) to account for between-center variability.

To explore functional relationships between preoperative anxiety or depression and pain, the model included categorical indicators for each summary scale (GAD-7 and PHQ-9) based on pre-defined cutoff points of each scale. The adjacent categories of depression and anxiety were collapsed if the parameter estimates of the adjacent categories consistently indicated no difference in 1-week postoperative pain across all pain measures. In summary, anxiety severity was grouped into none (0), minimal-mild (1–9), moderate (10–14), and severe (15–21) based on the GAD-7 score (possible range from 0 to 21). Depression severity was grouped into none (0), minimal (1–4), mild-moderate (5–14), moderately severe (15–19), and severe (20–27) based on PHQ-9 score (possible range from 0 to 27).

Results

Study Participants

The potential subject pool consisted of 2667 women. Table 1 lists the demographic characteristics of the study sample. The mean age was 49.7 (10.1) years old. The sample was predominantly white (87%) and non-Hispanic (95%). The majority of patients underwent TE/Implant reconstruction (69%) with DIEP flap reconstruction the next most frequent surgical procedure (17%). Most patients had immediate (93%) and bilateral (53%) reconstruction. The reason for mastectomy was cancer (therapeutic) for 90%. Of the 2667 study participants, 82.8% (N = 2207) had completed or partially completed PRO questionnaires at 1-week.

Table 1.

Study Sample Demographics (N=2667)

Variables	N	Percent
Age, mean (SD)	49.7 (10.1)
BMI, mean (SD)	26.7 (5.7)
Race
White	2317	87.8%
Black	171	6.5%
Asia	125	4.7%
Other	26	1.0%
Ethnicity
Non-Hispanic	2476	94.5%
Hispanic	144	5.5%
Laterality
Unilateral	1249	46.8%
Bilateral	1418	53.2
Type of Reconstruction
Tissue Expander/ Implant	1846	69.2%
Pedicle TRAM	94	3.5%
Free/MS TRAM	111	4.2%
DIEP	463	17.4%
Latissimus Dorsi	80	3.0%
SIEA	73	2.7%
Timing of Reconstruction
Immediate	2487	93.3%
Delayed	180	6.8%
Laterality
Unilateral	1249	46.8%
Bilateral	1418	53.2%
Reason for Mastectomy
Therapeutic	2392	89.7%
Prophylactic	275	10.3%
Comorbidities*
0	200	7.5%
1	2205	82.7%
2 or more	262	9.8%

^*Comorbidities include diagnosis of cancer

Acute Postoperative Pain

As expected, nearly all patients reported an increase in postoperative pain as reflected by the following change in scores: the MPQ-Sensory Rating from 3.1 (SD 4.5) to 10.5 (SD 6.3), the MPQ-Affective Rating from 1.6 (SD 2.0) to 2.9 (SD 2.3), the NPRS from 1.1 (SD 1.7) to 3.9 (SD 2.1) and the BREAST-Q Chest and Upper Body scale, for which lower scores indicate worse physical well-being and thus more pain, from 79.0 (SD 14.7) to 57.8 (SD 12.7).

Factors Associated with Acute Postoperative Pain

Younger age (P = .002), bilateral procedures (P < .0001), higher preoperative pain (P < .0001), preoperative anxiety (minimal-mild P = .01, moderate P < .01, severe P = .01) and preoperative depression (minimal P = .04, mild-moderate P < .01, moderate-severe P < .01, severe P = .01) were all significantly associated with greater acute postoperative pain as assessed by MPQ-Sensory pain scores (Table 2). Moreover, the degree of patients’ self-reported preoperative anxiety and depressive symptoms both appeared to bear a linear relationship with the magnitude of increased postoperative pain at 1-week on the MPQ-Sensory pain rating (see Figures 1 and 2 which portray the relationship between preoperative anxiety and depression scores, respectively, and postoperative pain ratings). Table 3 lists the regression analysis for the MPQ-Affective pain ratings. For this metric, younger age (P = .01), Asian race (P = .01), bilateral procedures (P = .01), prophylactic mastectomy (P < .01), preoperative pain (P < .0001), preoperative anxiety (minimal-mild P < .01, moderate P < .01, severe P < .0001) and preoperative depression (severe P < .01) were all related to increased MPQ-Affective pain scores. Younger age (P < .01), BMI (P = .04), bilateral procedures (P = .02), preoperative pain (P < .0001) preoperative anxiety (minimal-mild P = .05, moderate P < .01, severe P < .01) and depression (mild-moderate P = .02, severe P = .01)) were associated with higher NPRS scores at 1-week (Table 4).

Table 2.

Multiple Regression Analysis of Factors Associated with Increased Post-operative MPQ-Sensory Pain Ratings (N=1821)

Independent Variable	Coefficient	P-value
Age	−0.06	0.02
Race
White	Reference
Asian	−0.29	0.67
Black	0.26	0.66
Other	1.33	0.36
Ethnicity
Hispanic/Latino	0.02	0.97
BMI	0.01	0.76
Bilateral Procedure	1.59	<0.0001
Lymph Node Status
None	Reference
Sentinel Lymph Node Biopsy	−0.09	0.83
Axillary Lymph Node Dissection	0.26	0.59
Delayed Reconstruction	−0.51	0.46
Prophylactic Mastectomy	0.25	0.79
Preoperative Pain Score	0.36	<0.0001
Preoperative Anxiety
None	Reference
Minimal-Mild	1.09	0.01
Moderate	1.92	0.00
Severe	2.03	0.01
Preoperative Depression
None	Reference
Minimal	0.93	0.04
Mild-Moderate	1.59	0.00
Moderately Severe	3.50	0.00
Severe	3.99	0.01
Procedure Type
Tissue Expander/ Implant	Reference
DIEP	−1.20	0.01
Free/MS TRAM	0.26	0.72
Pedicle TRAM	−1.04	0.17
Latissimus Dorsi	0.35	0.69
SIEA	−2.41	0.02

Figure 1.

Relationship between severity of preoperative anxiety scores (GAD-7) and postoperative pain severity on the MPQ-Sensory Rating Scale

Figure 2.

Relationship between severity of preoperative depression scores (PHQ-9) and postoperative pain severity on the MPQ-Sensory Rating Scale

Table 3.

Multiple Regression Analysis of Factors Associated with Increased Post-operative MPQ-Affective Pain Scores (N=1748)

Independent Variable	Coefficient	P-value
Age	−0.01	0.01
Race
White	Reference
Asian	0.63	0.01
Black	0.10	0.64
Other	0.20	0.71
Ethnicity
Hispanic/Latino	0.32	0.17
BMI	0.00	0.81
Bilateral Procedure	0.29	0.01
Lymph Node Status
None	Reference
Sentinel Lymph Node Biopsy	0.07	0.67
Axillary Lymph Node Dissection	0.14	0.44
Delayed Reconstruction	0.00	1.00
Prophylactic Mastectomy	0.62	0.00
Preoperative Pain Score	0.28	<0.0001
Preoperative Anxiety
None	Reference
Minimal-Mild	0.54	0.00
Moderate	0.92	0.00
Severe	1.58	<0.0001
Preoperative Depression
None	Reference
Minimal	0.00	0.99
Mild-Moderate	0.31	0.11
Moderately Severe	0.52	0.16
Severe	1.75	0.00
Procedure Type
Tissue Expander/ Implant	Reference
DIEP	0.24	0.16
Free/MS TRAM	0.37	0.16
Pedicle TRAM	−0.01	0.98
Latissimus Dorsi	0.47	0.15
SIEA	−0.03	0.95

Table 4.

Multiple Regression Analysis of Factors Associated with Increased Post-operative NPRS Pain Scores (N=1872)

Independent Variable	Coefficient	P-value
Age	−0.02	0.00
Race
White	Reference
Asian	−0.06	0.81
Black	0.22	0.27
Other
Ethnicity
Hispanic/Latino	0.11	0.63
BMI	0.02	0.04
Bilateral Procedure	0.24	0.02
Lymph Node Status
None	Reference
Sentinel Lymph Node Biopsy	0.00	0.98
Axillary Lymph Node Dissection	0.03	0.85
Delayed Reconstruction	−0.32	0.17
Prophylactic Mastectomy	0.20	0.30
Preoperative Pain Score	0.28	<0.0001
Preoperative Anxiety
None	Reference
Minimal-Mild	0.29	0.05
Moderate	0.73	0.00
Severe	0.83	0.00
Preoperative Depression
None	Reference
Minimal	0.18	0.26
Mild-Moderate	0.40	0.02
Moderately Severe	0.40	0.24
Severe	1.33	0.01
Procedure Type
Tissue Expander/ Implant	Reference
DIEP	−0.18	0.25
Free/MS TRAM	−0.19	0.45
Pedicle TRAM	−0.72	0.01
Latissimus Dorsi	0.01	0.97
SIEA	0.21	0.51

Predictors of Condition-Specific Postoperative Pain

For the BREAST-Q Physical Well-Being scale used in this study, we found that younger age (P < .0001), bilateral procedures (P < .0001), immediate reconstruction (P < .0001), preoperative pain (P < .0001), preoperative anxiety (minimal-mild P = .03, moderate P < .01, severe P = .01) preoperative depression (mild-moderate P = .05, moderately severe P < .01, severe P < .01)) were significantly related to lower physical well-being scores at 1-week (Table 5).

Table 5.

Multiple Regression Analysis of Factors Associated with Increased BREAST-Q: Chest and Upper Body Scores (N=2207)

Independent Variable	Coefficient	P-value
Age	0.23	<0.0001
Race
White	Reference
Asian	2.25	0.06
Black	0.95	0.36
Other	0.88	0.74
Ethnicity
Hispanic/Latino	1.91	0.10
BMI	−0.02	0.62
Bilateral Procedure	−2.24	<0.0001
Lymph Node Status
None	Reference
Sentinel Lymph Node Biopsy	0.97	0.21
Axillary Lymph Node Dissection	1.34	0.11
Delayed Reconstruction	4.72	<0.0001
Prophylactic Mastectomy	0.47	0.63
Preoperative Pain Score	0.23	<0.0001
Preoperative Anxiety
None	Reference
Minimal-Mild	−1.69	0.03
Moderate	−4.17	0.00
Severe	−3.97	0.01
Preoperative Depression
None	Reference
Minimal	−0.65	0.42
Mild-Moderate	−1.77	0.05
Moderately Severe	−6.27	0.00
Severe	−7.63	0.00
Procedure Type
Tissue Expander/Implant	Reference
DIEP	4.89	<0.0001
Free/MS TRAM	2.51	0.06
Pedicle TRAM	6.71	<0.0001
Latissimus Dorsi	−0.90	0.56
SIEA	4.56	0.01

Comparison of Procedure Type and Postoperative Pain and Discomfort

The comparative effect of procedure type on postoperative pain showed variable results across our multiple pain scales. On the MPQ-Sensory scale, we found that patients with DIEP flaps reported less pain at one week (P < .01) as did those receiving SIEA reconstruction (P = .02) compared to patients with TE/Implant reconstruction. The other procedures did not show statistically significant differences compared to TE/Implant surgery. No procedural differences were detected on the MPQ-Affective scale. On the NPRS scale, only patients with PTRAMs reported less acute postoperative pain at one week compared to TE/Implants (P = .01). For the BREAST-Q, compared to TE/Implant surgery, the following reconstruction procedures were associated with less severe acute postoperative pain: PTRAM (P < .0001), DIEP (P < .0001), and SIEA (P = .01).

Discussion

This study explored factors associated with greater acute postoperative pain in a breast reconstruction cohort. We found consistent and strong association where younger age, bilateral reconstruction, and greater preoperative levels of pain, anxiety and depressive symptoms were associated with more severe acute postoperative pain across each measure of pain experience and a measure of condition-specific discomfort and physical dysfunction. These findings replicate previous evidence from major surgery studies that identify preoperative pain, mood disturbance and age as risk factors for greater acute postoperative pain. ^8,22,25,27 Our results also indicate that, relative to tissue expander/implant reconstruction, women who undergo DIEP, SIEA and pedicle TRAM surgery report significantly less postoperative pain and bodily discomfort. Moreover, we obtained suggestive evidence that immediate and prophylactic reconstructive surgery may be related to more severe discomfort and pain suffering, respectively. Of note, the data indicate no relation between lymph node status or BMI and acute postsurgical pain following breast reconstruction. The prospective methodology and large sample size are comparative strengths of this study. The findings have important implications for the identification of women who might be at risk for significant early postoperative pain following breast reconstruction, as acute postoperative pain is associated with poor functional outcomes, diminished quality of life and the risk for the development of persistent postsurgical pain.^6,8

Consistent with previous reports that assess both acute and more persistent postoperative pain following reconstruction, we found that our cohort of women receiving autologous flap reconstruction (DIEP, SIEA, PTRAM) experienced less early postoperative pain and morbidity compared to TE/Implant reconstruction. ^{9,20,35,37–39} Gassman et al. ¹⁹ describe features of TE/Implant reconstruction that are particularly related to acute postoperative pain and identified sustained expansion of musculocutaneous tissue as a possible explanation for the increased pain associated with tissue expander reconstruction. Extensiveness of surgery may also be an explanation for comparative increased acute post-reconstruction pain. For example, in our study women who underwent bilateral procedures consistently reported higher scores on all postoperative pain measures and greater site-specific physical discomfort. In addition, women receiving immediate rather than delayed reconstruction also reported more physical discomfort and dysfunction. Of note, we did not obtain any association between acute postoperative pain ratings with the addition of a sentinel lymph node biopsy or axillary lymph node dissection, although it is likely that lymph node surgery may not carry the same pain burden as mastectomy or breast reconstruction.

A methodological confound in our study and others that assess acute post-reconstruction pain concerns the degree to which mastectomy, reconstruction or some combination account for the observed pain outcomes. It is well established that persistent pain is an enduring clinical dilemma for many women who undergo mastectomy. ⁴⁰ The overwhelming majority of our sample underwent immediate reconstruction at the time of mastectomy, thus for these women reconstruction is nested within mastectomy at surgery. It has been generally assumed based on early accounts by Wallace et al^{. 9} and others ¹⁰ that breast reconstruction adds increased pain burden for women undergoing mastectomy, as the women in these studies with reconstruction reported more persistent pain than women receiving mastectomy alone. However, recently two retrospective cohort studies ^12,41 compared women who had undergone mastectomy alone or with immediate breast reconstruction, at least 6 months or longer from their surgery, for the presence of persistent pain. For both studies a range between 40–48% of the respective study samples reported some level of chronic pain. Neither study found evidence for increased risk of persistent pain with reconstruction compared to mastectomy alone, nor was type of breast reconstruction (e.g., TE/I or autologous reconstruction) associated with either the presence or intensity of chronic pain at follow-up. Thus, it is unclear the degree to which breast reconstruction may contribute to post-reconstruction pain beyond that attributable to mastectomy, and therefore it may be misleading when interpreting our findings on the surgical factors associated with postoperative pain to attribute the pain outcomes to reconstruction surgery alone.

Psychological disturbance has long been associated with greater pain experience,⁴² therefore it is not surprising that the degree of preoperative anxiety and depression in our study positively correlated with the magnitude of postoperative pain and replicates prior investigations.^25,27 Our evaluation of both anxiety and depression utilized an assessment tool that focuses on distress symptoms within the preceding two weeks. This method of evaluation identifies patients who are actively experiencing anxiety or depressive symptoms, which is of particular relevance for many women seeking immediate breast reconstruction following a recent diagnosis of breast cancer. ⁴³ Of clinical importance, we found that even mild levels of both anxiety and depressive symptoms consistently and significantly predicted more severe postoperative pain across the three pure pain measures. These correlations highlight the importance of evaluating anxiety and depressive symptoms at the preoperative clinic visit, even when these symptoms may be mild and potentially subclinical.

Preoperative pain is a known predictor of postoperative pain level, ¹⁸ and this finding was replicated in our breast reconstruction cohort. The mechanisms underlying the association between preoperative and postoperative pain is incompletely understood; however, researchers have theorized that the link may be either personality-dependent or neurologically predetermined. ⁴⁰ Patients who employ maladaptive pain coping strategies that are associated with enhanced pain experience (e.g., pain catastrophizing, pain-related fear) prior to surgery may be more likely to experience more severe postoperative pain.²⁶ Our study design did not include the assessment of pain-related cognitive coping variables, and they could have moderated the relationship between preoperative and postoperative pain obtained in the present analysis. Alternatively, greater postoperative pain in this cohort may be due to the effects of preoperative pain on central nervous system plasticity by way of the mechanism of central sensitization. Central sensitization may follow from a history of continuous noxious stimulation (e.g., preoperative pain) that enhances (postoperative) pain intensity as a result of a reduction in the threshold for activation and/or increased excitability of pain-related spinal dorsal horn neurons.⁴⁴

Younger age is associated with an increased likelihood of choosing breast reconstruction following mastectomy. ⁴⁵ Among our sample, younger patients reported higher levels of postoperative pain, and in particular reported more severe complaints of chest and upper body discomfort and reduced mobility. The relationship between age and postsurgical pain is an emerging finding, with recent studies demonstrating a predictive effect of younger age on increased postoperative pain.^8,22,23 It is possible that younger patients experience greater emotional distress in the context of illness that necessitates surgery compared to older patients, the latter of whom may see serious illness as an expected travail of aging. More specific to the present sample, it is understandable that a younger person may experience the diagnosis of either breast cancer or possessing a strong genetic predisposition to breast cancer with greater comparative emotional disturbance or perceived threat to future aspirations and quality of life. In addition, younger patients may also have greater concern for the aesthetic results and perceived altered body image that follows breast reconstruction, further amplifying the emotional effects of reconstruction. Such emotional upheaval, with accompanying symptoms of anxiety and depression, could intensify pain severity or undermine health behaviors that might mitigate pain symptoms.

Our study has several limitations. First, the treatment of postoperative pain was surgeon and institution-dependent. As our sample included multiple surgeons and multiple sites, it was not possible to standardize the postoperative pain treatment regimen. The differential treatment of post-surgical pain may have affected patients’ self-reported pain experience. While this is certainly a limitation, the inclusion of multiple surgeons and multiple centers also gives us a more heterogeneous sample of breast reconstruction patients. Our patient population is likely similar in composition to the broader breast reconstruction population, which increases the potential generalizability of our results. Second, while we attempted to standardize the implementation of the study protocol across all institutions, there was some variability in the timing of the preoperative assessment. While the timing of the postoperative survey was relatively precise (within a 7-day window), the preoperative survey was completed at the time of the patient’s initial visit. Based on the procedure process of each individual institution, some patients may have completed their preoperative surveys shortly after receiving a cancer diagnosis, while other patients may have completed the survey weeks or months after initial diagnosis. This may have influenced variability of patient anxiety and depression symptoms. Finally, any survey study has the potential to be biased by non-responders. While our response rate was high (83%), we do not have data on non-responders to consider how this group may differ from our sample population.

In conclusion, we found strong associations between younger age, bilateral reconstruction, TE/Implant reconstruction, preoperative pain and severity of both preoperative anxiety and depressive symptoms and greater acute postoperative pain following breast reconstruction. This information will be useful to plastic surgeons in identifying patients who may be at risk for greater postoperative pain. Once high-risk patients are identified, appropriate referrals can be considered to facilitate careful monitoring of postsurgical pain for selected patients. Early intervention can give clinicians the opportunity to reduce postoperative morbidity and improve patients’ surgical experience and satisfaction with breast reconstruction outcome.