Post-Discharge Nausea and Vomiting and Co-Occurring Symptoms in Women following Breast Cancer Surgery

Abstract

Purpose:

The purpose of this study was to investigate the influence of potential co-occurring symptoms, including fatigue, sleep disturbance, anxiety, depressive symptoms, and pain, on the incidence of post-discharge nausea (PDN) measured two days following discharge to home after surgery for breast cancer.

Design.

This study utilized a prospective, cross-sectional, observational design.

Methods

The sample was 334 women aged 27–88 years of age. Demographic data were collected from the patient and the medical record before surgery. Symptom data were collected 48 hours following surgery using the Patient Reported Outcome Measurement System (PROMIS^®) and numerical nausea and pain scales.

Findings

Eighty-five (25.4%) of study participants reported some nausea two days after discharge. Study participants who experienced PDN frequently described that it occurred after they left the hospital to drive home following their surgery. Unadjusted odds ratios showed the presence of co-occurring symptoms of anxiety, fatigue, sleep disturbance, and pain were all significantly associated with the presence of nausea 48 hours following surgery. Other significant factors associated with (PDN) were history of motion sickness, history of pregnancy-induced nausea, use of opioids, and type of surgery.

Conclusions

Same-day surgery nurses providing postoperative education for women following surgery for breast cancer should explain to patients that nausea may occur after they are discharged, especially those with known motion sickness. In addition, patients should be informed that other symptoms, especially fatigue, sleep disturbance, and anxiety, may co-occur.

Introduction and Background

The American Cancer Society (ACS) estimates that 287,850 women will be diagnosed with breast cancer in 2022¹, and the majority will be scheduled for surgical resection. Post anesthesia nurses know that women undergoing surgery for breast cancer are exposed to multiple triggers that may cause nausea and vomiting, including the type of surgical procedure, anesthesia during surgery, opioids for pain, co-occurring symptoms², and the disease itself.³ In addition, most surgical procedures for breast cancer are now performed as same-day surgery, using strict antiemetic protocols.⁴ Patients now report less immediate postoperative nausea and vomiting (PONV) than when hospital stays were extended but report higher levels of nausea and vomiting after being discharged.^5,6 This is known as post-discharge nausea and vomiting (PDNV) and is associated with most of the same risk factors as those observed with PONV.^{7, 8} Nausea and vomiting impact patient satisfaction and quality of life.⁹ PONV and PDNV, especially nausea, have been reported as more problematic than pain for women with breast cancer.^{10, 11, 5} The significant impact of nausea following surgery on quality of life in patients with cancer has been well described, most frequently accompanied by pain.¹²

Well-established risk factors for PONV and PDNV include female sex⁷, smoking status, a positive history of PONV/PDNV, opioid analgesia, and history of pregnancy-induced nausea and vomiting, and duration of anesthesia.¹³ Female biological sex is a stronger independent predictor than even a positive history of PONV.^14,15 It is not understood why female patients have an increased risk for nausea and vomiting, but this increased risk persists throughout life, even after menopause.¹⁶ Thus, PONV/PDNV can be considered a women’s health issue.^3,17 Ethnic differences in the experience and treatment of PONV have also been reported, with Black patients experiencing less nausea and vomiting than White patients.¹⁸ Studies from South Africa found that PONV was not a significant side effect in Black South Africans, and in fact, one study suggested a revised risk model for PONV, adding African ethnicity as a protector.^19,20

Though PONV has been successfully controlled, to a large extent, with new antiemetic drugs and disciplined use of enhanced recovery protocols,^{21 22} PDNV continues to be problematic due to a host of additional factors.^5,23 Patients scheduled for breast cancer surgery frequently experience high anxiety levels, which may also contribute to nausea and vomiting.^24,25 For many years, post-anesthesia care unit (PACU) nurses attributed PDNV to the oral opioids women were prescribed after discharge¹², with many women taking an oral pill before leaving the hospital. Pain requiring opioids for relief has long been considered the strongest co-occurring symptom with PDNV, but with the current shift away from the use of postoperative opioids,²⁶ the recognition that there are other causes of PDNV must be considered.

Very little research has focused on co-occurring symptoms in the first week following surgical treatment for breast cancer.^{27,28; 29} However, there is strong evidence establishing the clustering of pain, nausea, fatigue, and sleep disturbances from adjuvant therapy.^30–33 Thus, despite the increased attention given to nausea and vomiting, the recognition of major risk factors, and the success of enhanced recovery protocols, gaps remain, especially among patients discharged several hours after surgery.²¹ Moreover, with our increased understanding of the importance of how symptoms cluster or co-occur,^32,34 it is critical to describe the precise PDNV phenotype to help patients with breast cancer achieve the best possible outcomes. Therefore, this study aimed to investigate the influence of potential co-occurring symptoms, including fatigue, sleep disturbance, anxiety, depressive symptoms, and pain, on the incidence of nausea and vomiting measured two days following discharge to home after surgery.

Methods

Study Design

Though the data reported are from an ongoing longitudinal study, this study utilized a comparative, cross-sectional design focused on data collected 48 hours following surgery.

Study Participants and Setting

Patients scheduled for breast cancer surgery (without reconstruction) at UPMC Hillman Cancer Center affiliates, Magee-Womens Hospital, or UPMC East Hospital were recruited in the preoperative holding area a few hours before surgery. During the pandemic, patients were given flyers describing the study at their preoperative visits so that recruitment could occur by phone the week before surgery

Patients were eligible for study participation if they met the following criteria: 1) biological female aged 18–90 years, 2) diagnosed with early-stage breast cancer (Stage I, II, IIIA), 3) had a negative history of previous neurological conditions that could also be a cause of nausea and vomiting, 4) were given an American Society of Anesthesiologists (ASA) physical status of I, II, or III, 5) had access to a computer or a telephone to report symptoms experienced, and 6) were able to read and write in English.

Power Analysis

The final study sample size of 334 reported in this study was based on calculations for the ongoing longitudinal study to achieve 80% power, at the level of p< 0.05 and a moderate effect size of 0.357 to detect nausea-associated differences.

Procedures and Data Collection

The study was approved by the University of Pittsburgh Institutional Review Board. Clinically trained research associates received extensive instruction on participant recruitment and data collection procedures specific to the study. All study participants engaged in active, informed consent prior to data collection. Data collected before surgery included age, race/ethnicity, living status, history of PONV, history of pregnancy-induced nausea and vomiting, history of motion sickness, and smoking status. All postoperative medications, discharge prescriptions, PONV, and type of surgery were collected from the electronic health record. Forty-eight hours following discharge from the hospital, participants were called at home. During this phone call, they were asked to report the extent to which they had experienced nausea, vomiting, fatigue, sleep disturbance, anxiety, depressive symptoms, and pain as measured by the PROMIS^® tools described below. Participants were also asked if they had filled their prescription for pain medication, why they declined the prescription if they chose to do so, and how much of the prescription was taken if they chose to fill it. All data were entered into an online, HIPAA secure, Qualtrics survey approved by the University of Pittsburgh Institutional Review Board. Co-occurring symptoms were measured with the Patient-Reported Outcomes Measurement Information System (PROMIS^®)³⁵, a well-validated National Institutes of Health initiative to develop standardized items that measure feelings, function, and perceptions applicable to a range of conditions.³⁶ The post-discharge assessment consisted of the four nausea and vomiting questions from the validated PROMIS^® Gastrointestinal scale, ³⁷ a 0–10 numeric nausea scale,³⁸ and the PROMIS^®−29. The PROMIS^®−29 includes brief but valid assessments of functional status, anxiety, depressive symptoms, fatigue, pain interference, sleep disturbance, and social role.³⁹ An 11-point pain rating scale was also included. Norm-based scores have been developed for the PROMIS^®; a score of 50 represents the mean of the general population.⁴⁰ Using the PROMIS^®−29, a score above 50 indicates a worse outcome for anxiety, depression, sleep disturbance, and fatigue.⁴⁰ Additional qualitative comments recorded as call notes from study participants allowed for identifying symptom trends beyond the scope of the administered survey.

Therapeutic Agents

Anesthetic agents and opioids have long been known to be associated with PONV.⁴¹ Medications taken following discharge from the hospital were reported by the patient, and prescribed medications were confirmed by medical record review to assess medication adherence.

Data Analysis

All data were transferred from Qualtrics to Excel files for data cleaning and determination of missing data. SPSS, version 27. (IBM, Armonk, NY) was used for data analysis. All study participants responded to the 48-hour call; thus there was no missing data. Descriptive statistics were run to describe the sample. T-tests were used to compare mean age and pain scores between the group that experienced post-discharge nausea (PDN) and those who did not. Chi-Square analysis and Fisher Exact Tests were used to compare groups for variables measured categorically. Binary logistic regression was employed to determine the influence of co-occurring variables and other factors on the dependent variable of PDN.

Results

The sample for this study included 334 women with an average age of 60.0 ± 12.0 (range from 26–88) years. The majority of women were White (88%) and non-smokers (90%). Eighty-four (25.1%) of study participants reported some level of nausea two days after discharge. Though the majority of study participants (less than 10% as confirmed by chart audit) reported no nausea in the post-anesthesia care unit (PACU), many described that as soon as they were in the car for the drive home following the surgery, nausea occurred. Only 17 (5%) of study participants reported vomiting after discharge compared to the 85 (25.4%) who reported at least some nausea, with 40 (52%) reporting nausea as 4 or more on the 0–10point numeric nausea scale. Due to the low number of women who experienced vomiting, the analysis for this project focused on PDN. The comparison of study participants who experienced PDN or did not experience PDN is shown in Table 1. Pain scores were significantly different for the two groups (p<0.001), with the mean pain score for women with PDN 4.46 ± 2.32 on the 0–10-point scale and those women without PDN 2.82 ± 2.55. There was no significant difference when White women were compared to Black women; however, women who experienced PDN were significantly younger (p<0.007). More than half of the women who reported PDN had a positive history of motion sickness, but only 80 (35%) of the women who did not report PDN had a history of motion sickness. The use of opioids was also significantly associated with PDN (p<0.001).⁴² 159 of the 290 women who were prescribed opioids for use after discharge took at least one dose, 54 (33%) chose not to fill their prescription, and 77 filled their prescriptions but chose not to use or did not feel they were needed for management of postoperative pain.

Table 1.

Comparison of participants who experienced post-discharge nausea and those who did not (n=334).

	Experienced PDN (n=85)	Did not experience PDN (n=249)	p-value
Age (years)	56.94 ±11.9	61.16±11.8	p<0.00
Race - White (n= 269)	73 (25%)	226 (75%)	NS
Race - Black (n=30)	7 (9%)	23 (91%)	NS
Race - Asian (n=5)	3 (60%)	2 (40%)	NS
Positive History of Motion Sickness (n=131)	41(31%)	90 (69%)	p< 0.05
Positive History of Pregnancy Induced Nausea and Vomiting (n=128)	38 (45%)	90(36%)	p<0.067
Positive History of PONV (n=103)	28(30%)	65(70%)	p<0.85
Type of Surgery Segmental Mastectomy (n=232)	65(28%)	167(72%)
Type of Surgery Full Mastectomy or Bilateral Mastectomy (n=103)	20(23%)	83(86%)
Pain as measured on a 0–10 Numeric Scale	4.43 ± 2.32	2.81 ± 2.55	p<0.05
Use of Opioids after Discharge (at least one dose) (n=159)	53 (33%)	106 (67%)	p<0.01

Unadjusted odds ratios for predictive factors measured on the PROMIS^®29 and the presence vs. absence of post-discharge nausea are shown in Table 2. Anxiety, depressive symptoms, fatigue, sleep disturbance, and pain were all significantly associated with the presence of nausea.

Table 2.

Unadjusted odds ratios for co-occurring symptoms and PDN 48 hours after surgery. Individually, all co-occurring symptoms were statistically significant when analyzed separately.

Factor	B	SE	Wald	p-value	Odds Ratio	95% CI
Anxiety	0.41	0.133	31.74	p<0.003	1.041	1.01 – 1.07
Depressive symptoms	0.64	0.018	13.31	p< 0.000	1.066	1.03 – 1.10
Fatigue	0.206	0.36	33.26	p< 0.000	1.229	1.15 – 1.32
Sleep Disturbance	0.108	0.034	10.37	p< 0.001	1.11	1.04–1.90
Pain	0.240	0.053	20.83	p< 0.00	1.27	1.15 – 1.40

Correlations were assessed before calculating the adjusted odds ratio analysis with logistic regression. Due to the significant correlation between depressive symptoms and anxiety, only anxiety was included in the analysis. In addition, an assessment of multicollinearity found no areas of risk. The results of the adjusted binary regression analysis are shown in Table 3. The significant co-occurring symptoms were pain and fatigue. Other factors significantly associated with PDN in the model were the use of opioids, type of surgery, and history of motion sickness. Age and history of PONV were not significant factors associated with the presence of PDN in this group of women. Because of the very low number of current cigarette smokers in this sample (10%), smoking was not included as a factor in the final analysis

Table 3.

Binary Logistic regression results showing adjusted odds ratios.

				95% Confidence Levels
Variables	Wald	Sig (p)	B (OR)	Lower	Upper
Age	.785	0.270	0.988	0.96	1.10
History of Motion Sickness	4.25	0.01	1.859	1.03	3.35
Opioids for Postoperative Pain	4.731	0.030	1.973	1.07	3.64
Type of Surgery Segmental vs Total Mastectomy	4.095	0.022	2.342	1.12	4.86
Fatigue	17.084	0.003	1.147	1.04	1.13
Sleep Disturbance	0.767	0.363	1.017	0.98	1.05
Anxiety	0.767	0.361	1.014	0.98	1.05
Pain	1.533	0.05	1.138	0.97	1.30

Discussion

The purpose of this study was to determine the influence of fatigue, sleep disturbance, anxiety, depressive symptoms, and pain along with age and ethnicity on the incidence of nausea and vomiting measured two days following surgery. Fatigue and pain, as measured on the PROMIS^®29, were significantly associated with PDN. Sleep and anxiety were not significant co-occurring symptoms, inconsistent with other reports of women following breast cancer during adjuvant therapy.^29,43 However, previous studies did not collect data immediately or within a week following surgery and focused more on adjuvant therapy following surgery. Women in the immediate postoperative week naturally sleep more and may not yet experience anxiety and sleep disturbances as they recover from anesthesia. In addition, the anxiety from surgery would have subsided and they have not yet begun to worry about the next steps in their plan of care.

Twenty-five percent of study participants experienced PDN, which continues to be consistent with the 20–30% incidence reported by most researchers, ^7,44,45,46 regardless of interventions. One-night hospitalizations were the norm for much of the earlier PONV studies in women with breast cancer.⁴⁷ However, now that the majority of patients leave within several hours of surgery,⁶ the experience of PONV is often very different. Less than 10% of study participants reported the experience of nausea in the PACU, yet 25% experienced PDN, often as severe nausea in the drive home following surgery. This is consistent with the results of our preoperative assessment, indicating that more than 50% of study participants who experienced nausea also had a history of motion sickness.

With the opioid epidemic still a major worry, many women are discharged without prescriptions for pain, and many women who do receive prescriptions upon discharge do not fill them. Yet, the use of opioids continues to be a factor associated with nausea in women following breast cancer surgery. Five years ago, when women did not fill their post-discharge prescriptions, they would report they did not want to be “sick.”¹² Now, however, they are much more worried about “addiction” than nausea.⁴² Alsbrook and colleagues⁴⁸ reported recently that older women recovering from surgery for breast cancer were hesitant to ask their care team for stronger medications, even with pain levels of 10 on a 0–10 scale.⁴⁸ Inconsistent with our previous findings,^48,49 is that pain was significantly associated with post-discharge nausea when included in the regression model.

In this analysis, the surgical type was also significantly associated with the presence of nausea 48 hours following surgery. Women who were post-discharge following a total mastectomy or bilateral mastectomy experienced more PDN than those women who had a segmental mastectomy. The average time of surgery for a segmental mastectomy (or breast-conserving surgery) is approximately 70 minutes compared to 113 minutes for women having a total mastectomy, or 189 minutes for a bilateral mastectomy. More prolonged exposure to anesthetic agents is a well-established risk factor for PONV ^3, and thus would be expected in PDN as well.⁵⁰

A limitation of this study is the low diversity in the sample, with only 11% of the study sample being nonwhite. It is worth noting, however, that this percentage is representative of the demographics of Western Pennsylvania. In future studies, it will be imperative to increase the number of study participants that represent nonwhite populations. Another limitation is that data are patient reported outcomes, and not observed directly.

Clinical Implications

In conclusion, nurses providing postoperative education for women following surgery for breast cancer should recognize and explain to patients that nausea may occur after they are discharged. Assessment prior to discharge for women with more extensive surgery and with a positive history of motion sickness will identify those patients who need extra intervention; especially if they have a long drive to their homes post-discharge. This could include antiemetic medication in addition to the postoperative pain medication that may be required. Nurses caring for patients should also explain to patients who experience PDN that it is not unusual to have severe fatigue following surgery for breast cancer as well. Understanding the symptoms