Abstract
Background:
Cancer patients accept surgeries as part of their treatment. They may not be aware of the possibility of surgical pain persisting long after the surgery. Understanding chronic postsurgical pain is essential for effective pain management.
Aims:
We aimed to assess the prevalence of chronic postsurgical pain in cancer patients and the associated symptom burden.
Settings and Design:
This study was carried out at a tertiary cancer center. It was a cross-sectional study.
Materials and Methods:
Participants who underwent surgeries were asked to provide feedback on the MD Anderson Symptom Inventory at 3 months.
Statistical Analysis Used:
Descriptive statistics were used. Statistical tests included Kruskal–Wallis test, Chi-square test, Fisher's exact test, and Spearman's correlation. Logistic regression was used to assess the influence of variables on the presence or absence of chronic postsurgical pain.
Results:
Nine hundred and eighteen participants completed the study. Ninety-two percent (n = 840) were asymptomatic. Eight percent (n = 78) had postsurgical pain. Chronic postsurgical pain was influenced by the type of surgery (P = 0.01), specifically orthopedic and thoracic surgeries. Patients who receive epidurals are three times less likely to continue to have pain at 3 months.
Conclusions:
The prevalence of chronic postsurgical pain at 3 months in this study is lower than the rates in the literature. It is still associated with symptom burden that interferes with daily life. The risk of developing chronic postsurgical pain increases with thoracic and orthopedic surgeries. The risk may be lowered with epidural analgesia.
Keywords: Cancer patients, chronic postsurgical pain, persistent postsurgical pain, surgery
INTRODUCTION
Chronic postsurgical pain has been described as a major public issue.[1] Its prevalence ranges between 5% and 85%.[2] The rates depend on the type of surgery[1,2,3] and the measuring tool.[4]
Fifty percent of cancer patients experience pain because of the disease or its treatment.[5,6,7] In Jordan, 71% of cancer patients suffer pain that interferes with daily life.[8] Understanding chronic postsurgical pain, one prominent source in cancer patients, is essential to lessen suffering.
In this study, we aimed to assess the prevalence of chronic postsurgical pain in cancer patients and the associated symptom burden.
MATERIALS AND METHODS
The center's institutional review board (IRB, accredited by the Association for the Accreditation of Human Research Protection Programs) reviewed and approved the study (IRB# 13 KHCC 21, approved in 2013). We adhered to the IRB and the Helsinki Declaration (in 1975 and revised in 2013) mandates for research on human subjects. The IRB required that patients' informed consent be obtained in ample time before the surgery. Written informed consent was obtained from all patients participating in this study. This manuscript adheres to the applicable STROBE guidelines for reporting cross-sectional studies.
This was a cross-sectional research study. After 3 months of the surgery, we collected participants' feedback on their symptoms using the MD Anderson Symptom Inventory (MDASI). We ran this study between January 2013 and October 2016 at a single tertiary cancer center in Jordan.
The primary outcome was to determine the overall prevalence of chronic postsurgical pain in cancer patients. The secondary outcomes included assessing prevalence in different surgery types; exploring the impact of age, gender, and analgesia on the prevalence; and estimating symptom burden 3 months from surgery.
Selection criteria
Patients were selected if they met the selection criteria. A patient was eligible if found to be a Jordanian, 18 years or older, and had a planned surgery. The patient was excluded if found cognitively impaired, with a relapsed disease state, or had been treated for chronic pain.
During the preanesthesia clinic visit, the nurse coordinator or the anesthesiologist introduced patients to the study and, when possible, obtained their consent to participate. Patients' presurgical care, surgery, and postsurgical care followed the standard practices at the center. Three months following the surgery, the presence or absence of postsurgical pain was assessed. During a short phone call with the patient, the MDASI was completed.
We elected to use the MDASI as it offered several advantages. It can be easily collected during a phone call. It takes no more than 10 min to complete. It is available in Arabic, and patients can easily understand it. It also provides additional information on symptom severity and symptom interference with daily life. Finally, it can be applied broadly across cancer types and treatments.[9,10,11]
Participants were required to recall symptoms and intensity in the last 24 h.[12] We verified with the participants that the pain they described resulted from the surgery and no other reason. Data on the participants' characteristics and surgery were obtained from the patients' charts. Study data were collected into an access database until the analysis.
A patient was identified with chronic postsurgical pain if the pain from the surgery persisted a minimum of 3 months after the surgery.[13] Symptom intensity is “mild” if scores were >1 but <5 and “moderate to severe” if scores were ≥5.[14]
Sample size estimation
The minimum sample required was determined using different reported chronic postsurgical pain rates for leg amputation, thoracotomy, breast, and inguinal herniorrhaphy (60, 50, 30, and 10, respectively).[1,15] On average, 15–25 patients are admitted for surgery every week. At best, 60%–70% of patients will participate, and 10% may be lost to follow-up. Considering the recommendations for the precision proposed by Naing and Winn[16] and based on 95% confidence limits, different estimates were calculated using OpenEpi Copyright (c) 2003, 2008 Andrew G. Dean and Kevin M. Sullivan, Atlanta, GA, USA, an open-source calculator[17] [Table 1]. For a 10% prevalence with a 2% precision, 864 subjects were the minimum sample size necessary. With the required adjustment to ensure a complete sample, 960 subjects were required. Hence, we aimed to recruit 1000 participants.
Table 1.
Different sample size estimates
| Hypothesized prevalence (%) | Precision (%) | Sample size | Adjusted sample size |
|---|---|---|---|
| 10 | 2 | 864 | 960 |
| 30 | 5 | 323 | 359 |
| 50 | 5 | 384 | 427 |
| 60 | 5 | 369 | 410 |
A total of 1000 patients were screened and enrolled in the study. Eighty-two did not answer the phone. Nine hundred and eighteen participants completed the phone interviews and comprised the final analysis dataset.
We used descriptive statistics, counts, and percentages to report the study's results. When testing for significance, we used the Kruskal–Wallis test to assess the difference in age between groups. Chi-square and Fisher's exact tests were applied to the categorical variables. Spearman's correlation was used to evaluate the association between age and pain scores. Age was grouped into two distinct groups at a cutoff of 55 determined as the cutoff with the maximum difference with receiver operating characteristic curve analysis[18] (first group: ≥55 and second group: <55). The surgeries were grouped according to type.
MDASI mean scores with standard deviations (SDs) were calculated for total items (19 items), core symptom items (13 items), symptom interference items (6 items), and its subdimensions: affective subdimension (relations with others, enjoyment of life, and mood [REM]) and the activity subdimension (walking, general activity, and work [WAW]).
Logistic regression was used for the binary outcome of whether or not a participant is presented with chronic postsurgical pain at 3 months. Variables were selected when they were found statistically significant in univariate analysis. The model summary was reported.
Missing data were left as missing and were not substituted through imputation.
The significance level was determined at a P ≤ 0.05 for all analyses. The data analysis was done using R version 4.1.0 The R Foundation, Vienna, Austria (Camp Pontanezen, May 18, 2021) and appropriate packages through RStudio version 1.4.1717, The R Foundation, Vienna, Austria.[19,20,21,22,23]
RESULTS
Nine hundred and eighteen participants completed the study. The median age was 48 years, with an interquartile range between 40 and 58 years. Females represented two-thirds of the sample (n = 603, 66%). The majority of subjects were diagnosed with breast cancer (n = 378, 44%) and digestive/gastrointestinal cancers (n = 199, 22%). Breast and visceral surgeries were the most frequently performed surgeries, 38% (n = 353) and 24% (n = 221), respectively. Almost 69% (n = 632) of the surgeries took more than 3 h to complete, and those under 1 h were no more than 7% (n = 67). Two hundred and thirty-eight (26%) participants received epidural anesthesia. Epidural anesthesia is done intraoperatively and continues up to 5 days in most cases. It rarely extends to 6 or 7 days. Epidural drugs are a mix of bupivacaine 0.125% and fentanyl 2 μg.ml− 1. Only 27 (3%) were provided with patient-controlled analgesia (PCA), morphine, or fentanyl. Additional participant information is described in Table 2.
Table 2.
Characteristics of the participants
| Characteristic | Data value |
|---|---|
| Age | |
| Median (IQR) (years) | 48 (40-58) |
| Age≥55 years, n (%) | 285 (31) |
| Age<55 years, n (%) | 632 (69) |
| Missing, n (%) | 1 (0.1) |
| Gender, n (%) | |
| Females | 603 (66) |
| Males | 315 (34) |
| Primary diagnosis, n (%) | |
| Breast cancer | 378 (41) |
| Digestive/gastrointestinal cancers | 199 (22) |
| Endocrine and neuroendocrine cancers | 49 (5) |
| Genitourinary cancers | 75 (4) |
| Gynecologic cancers | 47 (5) |
| Head-and-neck cancers | 30 (3) |
| Hematologic/blood cancers | 18 (2) |
| Musculoskeletal cancers | 44 (5) |
| Neurologic cancers | 15 (2) |
| Respiratory/thoracic cancers | 26 (3) |
| Skin cancers | 6 (1) |
| Other cancers | 11 (1) |
| Missing | 20 (2) |
| Surgeries, n (%) | |
| Gynecology/obstetrics | 53 (6) |
| Visceral surgery | 221 (24) |
| Breast | 353 (38) |
| Urology | 80 (9) |
| ENT stomatology | 76 (8) |
| Orthopedic | 51 (6) |
| Thoracic | 57 (6) |
| Neurosurgery | 21 (2) |
| Others | 6 (1) |
| Surgery duration (h), n (%) | |
| <1 | 67 (7) |
| 1-2 | 219 (24) |
| >2 | 632 (69) |
| The subject had epidurals, n (%) | 238 (26) |
| The subject had PCA, n (%) | 27 (3) |
| The subject had epidural and PCA, n (%) | 1 (<1) |
| Subjects asymptomatic at 3 months, n (%) | 840 (92) |
| Subjects with chronic postsurgical pain at 3 months, n (%) | 78 (8) |
| Subjects with moderate-to-severe postsurgical pain at 3 months, n (%) | 27 (3) |
IQR=Interquartile range, ENT=Ear, nose, and throat, PCA=Patient- controlled analgesia
Three months after the surgery, 92% (n = 840) of the participants were asymptomatic (MDASI total score = 0). Eight percent (n = 78) had postsurgical pain, of which 35% (n = 27) complained of moderate-to-severe pain.
Table 3 lists the prevalence data for age, gender, and the different surgeries. The prevalence of chronic postsurgical pain was highest in thoracic and orthopedic surgeries, 20% (n = 10/51) and 16% (n = 9/57), respectively. The prevalence of chronic postsurgical pain was almost comparable between males and females (8% vs. 9%). The participants younger than 55 years had a slightly higher prevalence (9% vs. 7%) than those who were older (≥55 years).
Table 3.
Prevalence of chronic postsurgical pain at 3 months
| n | Prevalence, n (%) | |
|---|---|---|
| Overall | 918 | 78 (8) |
| Age (years) | ||
| Age<55 | 633 | 58 (9) |
| Age≥55 | 285 | 20 (7) |
| Gender | ||
| Male | 315 | 25 (8) |
| Female | 603 | 53 (9) |
| Surgery type | ||
| Gynecology/obstetrics | 53 | 4 (8) |
| Visceral surgery | 221 | 13 (6) |
| Breast | 353 | 35 (10) |
| Urology | 80 | 5 (6) |
| ENT stomatology | 76 | 2 (3) |
| Orthopedic | 51 | 10 (20) |
| Thoracic | 57 | 9 (16) |
| Neurosurgery | 21 | 0 |
| Others | 6 | 0 |
| Selected surgeries | ||
| Mastectomy | 213 | 18 (8) |
| Axillary dissection | 111 | 9 (8) |
| Hysterectomy | 27 | 2 (7) |
ENT=Ear, nose, and throat
MDASI per item analysis is described in Table 4. Overall, 8% (n = 78) of the participants suffered postsurgical pain at 3 months. Six percent experienced mild pain (n = 51), and moderate-to-severe pain was experienced by 3% (n = 27). Fatigued and numbness/tingling were the second-highest experienced symptoms (6%). Both correlated with the participants' pain scores (Spearman's correlation = 0.85). Four percent (n = 41) indicated that symptoms interfered with their daily lives. Interference with the general activity and work was most reported (6% of participants). Aside from pain, individuals that experienced moderate-to-severe symptoms did not exceed 1%.
Table 4.
MD Anderson Symptom Inventory item analysis
| Symptom intensity | Mean score (SD) | Association with pain score | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| Mild (1≤score<5), n (%) | Moderate to severe (score≥5), n (%) | All subjects | Subjects with chronic postsurgical pain | Spearman’s correlation | P | |
| Core symptom items | ||||||
| Pain | 51 (6) | 27 (3) | 0.35 (1.21) | 4.06 (1.41) | - | - |
| Fatigue | 48 (5) | 8 (<1) | 0.18 (0.85) | 2.11 (2.13) | 0.85 | <0.001 |
| Nausea | 10 (1) | 3 (<1) | 0.04 (0.42) | 0.49 (1.37) | 0.41 | <0.001 |
| Disturbed sleep | 34 (4) | 4 (<1) | 0.12 (0.63) | 1.42 (1.69) | 0.70 | <0.001 |
| Distressed (upset) | 25 (3) | 5 (<1) | 0.08 (0.53) | 0.99 (1.55) | 0.62 | <0.001 |
| Shortness of breath | 4 (<1) | 0 | 0.01 (0.14) | 0.10 (0.47) | 0.22 | <0.001 |
| Remembering things | 3 (<1) | 3 (<1) | 0.02 (0.32) | 0.27 (1.09) | 0.27 | <0.001 |
| Lack of appetite | 12 (1) | 5 (<1) | 0.06 (0.48) | 0.67 (1.52) | 0.46 | <0.001 |
| Drowsy (sleepy) | 8 (<1) | 2 (<1) | 0.03 (0.35) | 0.37 (1.14) | 0.36 | <0.001 |
| Dry mouth | 5 (<1) | 3 (<1) | 0.03 (0.41) | 0.36 (1.39) | 0.31 | <0.001 |
| Sad | 13 (1) | 1 (<1) | 0.03 (0.29) | 0.38 (0.93) | 0.43 | <0.001 |
| Vomiting | 4 (<1) | 2 (<1) | 0.02 (0.33) | 0.27 (1.12) | 0.28 | <0.001 |
| Numbness or tingling | 49 (5) | 8 (<1) | 0.18 (0.81) | 2.17 (1.85) | 0.85 | <0.001 |
| Interference items | ||||||
| General activity | 46 (5) | 13 (1) | 0.21 (0.88) | 2.45 (1.91) | 0.87 | <0.001 |
| Mood | 26 (3) | 6 (<1) | 0.10 (0.58) | 1.14 (1.66) | 0.65 | <0.001 |
| Work (including work around the house) | 40 (4) | 12 (1) | 0.19 (0.90) | 2.27 (2.21) | 0.81 | <0.001 |
| Relations with other people | 9 (<1) | 1 (<1) | 0.03 (0.30) | 0.33 (0.99) | 0.36 | <0.001 |
| Walking | 10 (1) | 5 (<1) | 0.06 (0.52) | 0.68 (1.66) | 0.43 | <0.001 |
| Enjoyment of life | 19 (2) | 1 (<1) | 0.05 (0.43) | 0.58 (1.39) | 0.51 | <0.001 |
| Summary | ||||||
| MDASI overall | 33 (4) | 1 (<1) | 0.09 (0.41) | 1.11 (0.95) | - | - |
| MDASI symptom items | 29 (3) | 1 (<1) | 0.09 (0.40) | 1.05 (0.91) | - | - |
| MDASI interference items | 40 (4) | 1 (<1) | 0.11 (0.49) | 1.24 (1.22) | 0.88 | <0.001 |
| MDASI WAW | 45 (5) | 7 (<1) | 0.15 (0.70) | 1.80 (1.70) | 0.88 | <0.001 |
| MDASI REM | 23 (3) | 1 (<1) | 0.06 (0.37) | 0.68 (1.07) | 0.68 | <0.001 |
MDASI=MD Anderson Symptom Inventory, WAW=Walking, general activity, and work, REM=Relations with others, enjoyment of life, and mood, SD=Standard deviation
The mean scores of MDASI for subjects with chronic postsurgical pain are described in Table 4. Participants' distribution for each symptom and interference item with respect to their intensity is depicted in Figures 1 and 2. In general, symptoms were mild (mean scores <5). The pain item score mean was 4.06 (SD = 1.41). Fatigue and numbness/tingling means were 2.11 (SD = 2.13) and 2.17 (SD = 1.85), respectively. The general activity score mean was 2.45 (SD = 1.91), and work had a mean of 2.27 (SD = 2.21). The remaining symptoms had mean scores below 1 or 2. Overall, there is a marked shift in scores of MDASI core symptom, interference, REM, and WAW items between mild- and moderate-to-severe pain (P < 0.001 in all) [Figure 3]. There were no gender differences in MDASI scores for core symptom, interference, REM, and WAW items (P = 0.64, 0.83, 0.98, and 0.92, respectively) [Figure 4]. Similarly, age showed no significant differences in MDASI scores for core symptom, interference, REM, and WAW items (P = 0.24, 0.25, 0.14, and 0.29, respectively. Fifty-five years was used as the cutoff) [Figure 5].
Figure 1.
The distribution of participants based on the symptom experienced and the intensity at 3 months. MDASI = MD Anderson Symptom Inventory
Figure 2.
The distribution of participants based on the interference experienced and the intensity at 3 months. MDASI = MD Anderson Symptom Inventory
Figure 3.
Boxplot of MDASI; core item, interference item, REM item, and WAW item scores, considering the intensity of chronic postsurgical pain at 3 months. MDASI = MD Anderson Symptom Inventory, REM = Relations with others, enjoyment of life, and mood, WAW = Walking, general activity, and work
Figure 4.
Boxplot of MDASI; core item, interference item, REM item, and WAW item scores, considering the participants' gender. MDASI = MD Anderson Symptom Inventory, REM = Relations with others, enjoyment of life, and mood, WAW = Walking, general activity, and work
Figure 5.
Boxplot of MDASI; core item, interference item, REM item, and WAW item scores, considering the participants' age. MDASI = MD Anderson Symptom Inventory, REM = Relations with others, enjoyment of life, and mood, WAW = Walking, general activity, and work
The comparison between participants with and without chronic postsurgical pain regarding the factors age, gender, surgery type and duration, and use of epidurals or PCA is shown in Table 5. Chronic postsurgical pain was influenced by the type of surgery (P = 0.01), specifically orthopedic and thoracic surgeries (P = 0.008 and 0.0499, respectively). Differences in age or gender were statistically insignificant (P = 0.28 and 0.66, respectively). Age did not correlate with pain scores even when grouped by gender (Spearman's rank correlation was ‒0.2678 and ‒0.0876 for males and females, respectively) [Table 6 and Figure 6]. The duration of surgery, including surgeries that took more than 3 h to complete, did not show any meaningful statistical difference (P = 0.49).
Table 5.
Comparison of the characteristics of participants, those without and those with chronic postsurgical pain at 3 months
| Without chronic postsurgical pain (n=840) | With chronic postsurgical pain (n=78) | P* | OR (95% CI) | |
|---|---|---|---|---|
| Age (years) | ||||
| Median (IQR) | 48 (40-58) | 47 (39-55) | 0.26 | 0.99 (0.97-1.01)† |
| Age<55, n (%) | 575 (68) | 58 (74) | ||
| Age≥55, n (%) | 265 (32) | 20 (26) | 0.28 | 0.75 (0.44-1.27) |
| Gender, n (%) | ||||
| Male | 290 (35) | 25 (32) | ||
| Female | 550 (65) | 53 (68) | 0.66 | 1.12 (0.68-1.84) |
| Surgery type, n (%) | ||||
| Gynecology/obstetrics | 49 (6) | 4 (5) | >0.99 | 0.87 (0.31-2.49) |
| Visceral surgery | 208 (25) | 13 (17) | 0.11 | 0.61 (0.33-1.12) |
| Breast | 318 (38) | 35 (45) | 0.27 | 1.34 (0.84-2.13) |
| Urology | 75 (9) | 5 (6) | 0.67 | 0.70 (0.27-1.78) |
| ENT stomatology | 74 (9) | 2 (3) | 0.053 | 0.27 (0.07-1.13) |
| Orthopedic | 41 (5) | 10 (13) | 0.008 | 2.87 (1.38-5.97) |
| Thoracic | 48 (6) | 9 (12) | 0.0499 | 2.15 (1.01-4.57) |
| Neurosurgery | 21 (3) | 0 | 0.25 | 0.24 (0.01-4.04)‡ |
| Others | 6 (<1) | 0 | >0.99 | 0.82 (0.05-14.65)‡ |
| Selected surgeries, n (%) | ||||
| Mastectomy | 195 (23) | 18 (23) | >0.99 | 0.99 (0.57-1.72) |
| Axillary dissection | 102 (12) | 9 (12) | >0.99 | 0.94 (0.46-1.95) |
| Hysterectomy | 25 (3) | 2 (3) | >0.99 | |
| Surgery duration (h), n (%) | ||||
| <1 | 61 (7) | 6 (8) | 0.82 | 1.06 (0.44-2.55) |
| 1-2 | 198 (24) | 21 (27) | 0.51 | 1.19 (0.71-2.02) |
| ≥3 | 581 (69) | 51 (65) | 0.49 | 0.84 (0.52-1.37) |
| Epidural, n (%) | 230 (27) | 8 (10) | <0.001 | 0.30 (0.14-0.64) |
| PCA, n (%) | 24 (3) | 3 (4) | 0.49 | 1.36 (0.40-4.62) |
*P≤0.05 is significant, †Odds obtained through logistic regression with one variable, ‡Haldane-Anscombe correction applied. CI=Confidence interval, PCA=Patient-controlled analgesia, OR=Odds ratio, ENT=Ear, nose, and throat, IQR=Interquartile range
Table 6.
Correlation between age and pain scores at 3 months
| Group | n | Spearman’s rank correlation rho | 95% confidence limits | P* | |
|---|---|---|---|---|---|
|
| |||||
| LL | UL | ||||
| All participants | 918 | −0.0397 | −0.1046 | 0.0246 | 0.2295 |
| Participants with pain at 3 months | 78 | −0.2028 | −0.4059 | 0.0219 | 0.0749 |
| Males with pain at 3 months | 25 | −0.2678 | −0.5959 | 0.1478 | 0.1955 |
| Females with pain at 3 months | 53 | −0.0876 | −0.3488 | 0.1880 | 0.533 |
*P≤0.05 is significant. LL=Lower limit, UL=Upper Limit
Figure 6.
Scatter plot of MDASI pain item score against age for males and females. MDASI = MD Anderson Symptom Inventory
The use of epidurals had a statistically significant difference compared to the use of PCA (P < 0.001 vs. 0.49). Participants who had epidurals were three times less likely to have persistent pain at 3 months (odds ratio: 0.3, with a 95% confidence limit between 0.14 and 0.64). A patient is three times as likely to continue to suffer from pain at 3 months when the patient was subject to orthopedic surgery, holding the other factors constant [Table 7]. The patient would have persistent pain at 3 months, twice as likely if subject to thoracic surgery. Under the same model, patients who receive epidurals are three times less likely to continue to have pain at 3 months.
Table 7.
Summary data of a logistic model for chronic postsurgical pain at 3 months fitted against orthopedic and thoracic surgeries and the use of epidural anesthesia
| Parameter | Estimate | SE | Z | P | OR | 95% confidence limits | |
|---|---|---|---|---|---|---|---|
|
| |||||||
| LL | UL | ||||||
| Intercept | −2.3325 | 0.1413 | −16.5120 | < 0.001 | 0.10 | 0.07 | 0.13 |
| Orthopedic surgery | 1.2365 | 0.3854 | 3.2081 | 0.001 | 3.44 | 1.54 | 7.11 |
| Thoracic surgery | 0.8929 | 0.3921 | 2.2774 | 0.02 | 2.44 | 1.07 | 5.06 |
| Epidural anesthesia | −1.2517 | 0.3842 | −3.2576 | 0.001 | 0.29 | 0.12 | 0.57 |
LL=Lower limit, UL=Upper limit, OR=Odds ratio, SE=Standard error
Similarly, we compared the presence or absence of moderate-to-severe postsurgical pain concerning the factors shown in Table 8. Age may contribute to persistent moderate-to-severe pain at 3 months (P = 0.0098). This association did not hold when age was fitted with orthopedic surgery in a logistic model [Table 9]. At 3 months, orthopedic surgery was the only surgery associated with moderate-to-severe pain. No other factors showed a significant association.
Table 8.
Comparison of the characteristics of participants, those without and those with moderate-to-severe postsurgical pain at 3 months
| Without moderate-to-severe pain (n=891) | With moderate-to-severe pain (n=27) | P* | Odds (95% CI) | |
|---|---|---|---|---|
| Age (years) | ||||
| Median (IQR) | 48 (40-58) | 43 (32-49.5) | 0.0098 | 0.96 (0.93-0.99) |
| Age <55, n (%) | 609 (68) | 24 (89) | ||
| Age ≥55, n (%) | 282 (32) | 3 (11) | 0.0206 | 0.27 (0.08-0.90) |
| Gender, n (%) | ||||
| Male | 304 (34) | 11 (41) | ||
| Female | 587 (66) | 16 (59) | 0.4752 | 0.75 (0.35-1.64) |
| Surgery type, n (%) | ||||
| Gynecology/obstetrics | 52 (6) | 1 (4) | >0.99 | 0.62 (0.08-4.66) |
| Visceral surgery | 218 (24) | 3 (11) | 0.1673 | 0.39 (0.12-1.29) |
| Breast | 341 (38) | 12 (44) | 0.516 | 1.29 (0.60-2.79) |
| Urology | 79 (9) | 1 (4) | 0.5035 | 0.40 (0.05-2.95) |
| ENT stomatology | 74 (8) | 2 (7) | >0.99 | 0.88 (0.21-3.80) |
| Orthopedic | 43 (5) | 8 (30) | <0.001 | 8.30 (3.44-20.04) |
| Thoracic | 57 (6) | 0 | 0.4048 | 0.26 (0.02-4.38)† |
| Neurosurgery | 21 (2) | 0 | >0.99 | 0.74 (0.04-12.47)† |
| Others | 6 (<1) | 0 | >0.99 | 2.48 (0.14-45.08)† |
| Selected surgeries, n (%) | ||||
| Mastectomy | 207 (23) | 6 (22) | >0.99 | 0.94 (0.38-2.37) |
| Axillary dissection | 108 (12) | 3 (11) | >0.99 | 0.91 (0.27-3.06) |
| Hysterectomy | 26 (3) | 1 (4) | 0.5586 | 1.28 (0.17-9.79) |
| Surgery duration (h), n (%) | ||||
| <1 | 66 (7) | 1 (4) | 0.7146 | 0.48 (0.06-3.60) |
| 1-2 | 212 (24) | 7 (26) | 0.8192 | 1.12 (0.47-2.69) |
| ≥3 | 613 (69) | 19 (70) | >0.99 | 1.08 (0.47-2.49) |
| Epidural, n (%) | 233 (26) | 5 (19) | 0.5046 | 0.64 (0.24-1.71) |
| PCA, n (%) | 27 (3) | 0 | >0.99 | 0.57 (0.03-9.61)† |
*P≤0.05 is significant, †Haldane-Anscombe correction applied. CI=Confidence interval, PCA=Patient-controlled analgesia, ENT=Ear, nose, and throat, IQR=Interquartile range
Table 9.
Summary data of a logistic model for postsurgical moderate-to-severe pain at 3 months fitted against age and orthopedic surgery
| Parameter | Estimate | SE | Z value | P* | OR | 95% confidence limits | |
|---|---|---|---|---|---|---|---|
|
| |||||||
| LL | UL | ||||||
| Intercept | −3.5525 | 0.2483 | −14.3075 | 0.0000 | 0.03 | 0.02 | 0.05 |
| Age ≥55 | −1.1058 | 0.6240 | −1.7723 | 0.0763 | 0.33 | 0.08 | 0.98 |
| Orthopedic surgery | 1.9622 | 0.4545 | 4.3178 | 0.0000 | 7.12 | 2.78 | 16.86 |
*P≤0.05 is significant. LL=Lower limit, UL=Upper limit, SE=Standard error, OR=Odds ratio
DISCUSSION
Eight percent of the study participants continued to have postsurgical pain at 3 months. This was well within reported rates of 5%–85%.[2] However, the rates we estimated are low compared to most literature reports. For example, breast surgeries in our study constituted 38% of the total surgeries. Mastectomies (n = 213) made up 60% of the breast surgeries and 23% of the total. Postsurgical pain following mastectomies is reported between 20 and 50%.[1,3] In Beloeil and Sulpice account, the rates are between 10% and 60%.[24] Our study's rate for postsurgical pain following mastectomy is only 8%.
Several reasons can help explain having low rates. Prevalence rates for postsurgical pain are very much dependent on the study design and tool used in the assessment.[4] Different studies measure pain prevalence at different time points (24 h, 3, 6, 12, or 24 months), and it is a challenge to match studies. The definition of the chronicity of postsurgical pain has been subject to debate.[25]
Patients may be reluctant to report pain for various reasons.[15,26] Patients were said to dismiss pain when thinking about their primary condition.[27] Patients have voiced concerns about opioids and the possibility of addiction.[15,28,29] Patients want to be good for their physicians.[26] Older patients do not want to be a nuisance to their families or caregivers. They are also likely to have difficulty in communicating pain to their physicians.[29]
How a given culture perceives pain influences the willingness to express pain.[28] Arabs have a tendency to underreport pain because they believe that it is an act of God and are willing to bear the pain until they no longer tolerate it.[24,30] Male Arabs are often discouraged from expressing pain because it is a sign of weakness.[30]
The quality of care for cancer patients varies between the hospitals that offer the service in Jordan.[31] This may influence the recovery of patients and eventually the experience of pain.
Gender and age differences have been discussed in the literature. Akkaya and Ozkan[3] have discussed reports showing demographic differences for old versus young age or women versus men. A recent review by Mills et al.[32] lists age and gender as factors in developing chronic postsurgical pain. This has not been the case in our study, even when selecting specific surgeries to explore (P > 0.05). Such differences may be evident when a study is designed and controlled to focus on a particular surgery. It may be the case when developing moderate-to-severe chronic postsurgical pain rather than mild. In our study, age was associated with moderate-to-severe postsurgical pain at 3 months. Still, we ruled it out as a factor in the logistic regression model. Beyond that, we are not aware of any other plausible explanation.
Our results show that the different surgeries have different prevalence rates for chronic pain. Orthopedic and thoracic surgeries were significant predictive factors. There may indeed be a need to inform patients about the possibility of developing chronic pain in similar surgeries, as was suggested by Choinière et al.[27]
Peters et al.[33] reported that patients undergoing surgeries that lasted longer than 3 h were at risk for chronic postsurgical pain at 6 months. Our study showed that duration did not associate with chronic postsurgical pain at 3 months. This suggests that time may not adequately reflect the complexity of the operation and the surgical techniques employed to help predict developing chronic postsurgical pain.
Bouman et al.[34] and Park et al.[35] discussed that epidural analgesia may reduce the risk of chronic postsurgical pain. Similarly, our results suggest that epidural analgesia can reduce the chances of developing postsurgical pain.
The MDASI item scores show that the symptom burden at 3 months is generally mild. However, shifting toward individuals with moderate-to-severe pain, the symptom burden increases. This supports the point that chronic postsurgical pain is a burdensome complication. Uncontrolled pain increases the general symptom burden, impacting the quality of life and affecting the patient's adherence to therapy and satisfaction with care.[7,36]
Although based on a large sample, the study results should be reviewed with caution. The majority of patients were subject to breast and visceral surgeries. It is worth noting that breast cancer and colorectal cancer are the highest reported cancers in the country.[37] Ninety-two percent had an MDASI total score of 0. If patients refrained from informing the surveyors about their actual pain status, we do not know. Confounding bias may exist because factors such as the disease stage and status, body mass index, social characteristic, or performance status were not collected and accounted for. The study assessed the point prevalence of chronic postsurgical pain. A 24-h recall will tell us about that single point in time. Patients may be forgetful or feeling well or even bad at that time, which may cloud their perception of pain. As it is point prevalence, we have no other information if this pain carried over the 3 months or was just a breakthrough episode of acute pain.
We worked with the International Association for the Study of Pain (IASP) definition[13] and a short recall of 24 h. However, pain is a complex phenomenon and is not a constant. Different factors can influence the perception of pain, including heredity.[1,38] Patients may not be in pain at that moment in time when they were asked about their pain.
The study results provide a glimpse of chronic postsurgical pain in cancer patients undergoing different surgeries. Furthermore, it includes information on the symptom burden associated with chronic postsurgical pain. Future research may focus on specific surgeries, surgery-related factors, and the appropriate patient characteristics (e.g., socioeconomic, medical, and demographic). A better understanding of the burden of chronic postsurgical pain on patients is possible if prevalence is assessed based on duration rather than a single point in time. Three months by definition is sufficient, but following up with patients for more extended periods may be more appropriate.
CONCLUSIONS
The prevalence of chronic postsurgical pain at 3 months in this study is lower than the rates in the literature. Nevertheless, it is still associated with symptom burden that interferes with daily life. The risk of developing chronic postsurgical pain increases with thoracic and orthopedic surgeries. The risk may be lowered with epidural analgesia.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Macrae WA. Chronic post-surgical pain: 10 years on. Br J Anaesth. 2008;101:77–86. doi: 10.1093/bja/aen099. [DOI] [PubMed] [Google Scholar]
- 2.Hetmann F, Kongsgaard UE, Sandvik L, Schou-Bredal I. Prevalence and predictors of persistent post-surgical pain 12 months after thoracotomy. Acta Anaesthesiol Scand. 2015;59:740–8. doi: 10.1111/aas.12532. [DOI] [PubMed] [Google Scholar]
- 3.Akkaya T, Ozkan D. Chronic post-surgical pain. Agri. 2009;21:1–9. [PubMed] [Google Scholar]
- 4.Drosos GI, Triantafilidou T, Ververidis A, Agelopoulou C, Vogiatzaki T, Kazakos K. Persistent post-surgical pain and neuropathic pain after total knee replacement. World J Orthop. 2015;6:528–36. doi: 10.5312/wjo.v6.i7.528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Haumann J, Joosten EB, Everdingen MH. Pain prevalence in cancer patients: Status quo or opportunities for improvement? Curr Opin Support Palliat Care. 2017;11:99–104. doi: 10.1097/SPC.0000000000000261. [DOI] [PubMed] [Google Scholar]
- 6.Marcus DA. Epidemiology of cancer pain. Curr Pain Headache Rep. 2011;15:231–4. doi: 10.1007/s11916-011-0208-0. [DOI] [PubMed] [Google Scholar]
- 7.Neufeld NJ, Elnahal SM, Alvarez RH. Cancer pain: A review of epidemiology, clinical quality and value impact. Future Oncol. 2017;13:833–41. doi: 10.2217/fon-2016-0423. [DOI] [PubMed] [Google Scholar]
- 8.Alawneh A, Anshasi H, Khirfan G, Yaseen H, Quran A. Symptom prevalence of patients with cancer in a tertiary cancer center in Jordan. Gulf J Oncolog. 2017;1:37–43. [PubMed] [Google Scholar]
- 9.Nabila R, Mimoun Z. Arabic tools for assessment of multidimensions of pain and discomfort related to cancer. Asian Pac J Cancer Prev. 2016;17:2619–24. [PubMed] [Google Scholar]
- 10.The University of Texas MD Anderson Cancer Center. The MD Anderson Symptom Inventory. 2022. [Last accessed on 2021 Nov 18]. Available from: https://www.mdanderson.org/research/departments-labs-institutes/departmentsdivisions/symptom-research/symptom-assessment-tools/md-andersonsymptom-inventory.html .
- 11.Nejmi M, Wang XS, Mendoza TR, Gning I, Cleeland CS. Validation and application of the Arabic version of the M. D. Anderson symptom inventory in Moroccan patients with cancer. J Pain Symptom Manage. 2010;40:75–86. doi: 10.1016/j.jpainsymman.2009.12.007. [DOI] [PubMed] [Google Scholar]
- 12.Shi Q, Trask PC, Wang XS, Mendoza TR, Apraku WA, Malekifar M, et al. Does recall period have an effect on cancer patients' ratings of the severity of multiple symptoms? J Pain Symptom Manage. 2010;40:191–9. doi: 10.1016/j.jpainsymman.2009.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schug SA, Lavand'homme P, Barke A, Korwisi B, Rief W, Treede RD, et al. The IASP classification of chronic pain for ICD-11: Chronic postsurgical or posttraumatic pain. Pain. 2019;160:45–52. doi: 10.1097/j.pain.0000000000001413. [DOI] [PubMed] [Google Scholar]
- 14.Cleeland C. The M. D. Anderson Symptom Inventory User Guide. The University of Texas MD Anderson Cancer Center. 2016:1–83. [Google Scholar]
- 15.Kwon JH. Overcoming barriers in cancer pain management. J Clin Oncol. 2014;32:1727–33. doi: 10.1200/JCO.2013.52.4827. [DOI] [PubMed] [Google Scholar]
- 16.Naing L, Winn T, Nordin R. Pratical issues in calculating the sample size for prevalence studies. Arch Orofacial Sci. 2006;1:9–14. [Google Scholar]
- 17.Dean AG, Sullivan KM, Soe MM. OpenEpi Open Source Epidemiologic Statistics for Public Health. Emory University, Rollins School of Public Health; open source software for epidemiologic statistics. [Last accessed on 2012 Dec 13]. Available from: www.OpenEpi.com .
- 18.Sing T, Sander O, Beerenwinkel N, Lengauer T. ROCR: Visualizing classifier performance in R. Bioinformatics. 2005;21:3940–1. doi: 10.1093/bioinformatics/bti623. [DOI] [PubMed] [Google Scholar]
- 19.Lüdecke D. _sjPlot: Data Visualization for Statistics in Social Science. 2018. [Last accessed on 2022 Mar 22]. Available from: https://CRAN.R-project.org/package=sjPlot .
- 20.Wickham H, François R, Henry L, Müller K. dplyr: A Grammar of Data Manipulation. 2018. [Last accessed on 2022 Mar 22]. Available from: https://CRAN.R-project.org/package=dplyr .
- 21.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2021. [Last accessed on 2022 Mar 22]. Camp Pontanezen. Available from: https://www.R-project.org/ [Google Scholar]
- 22.Grolemund G, Wickham H. Dates and times made easy with lubridate. J Stat Softw. 2011;40:1–25. [Google Scholar]
- 23.Wickham H, Averick M, Bryan J, Chang W, McGowan L, François R, et al. Welcome to the tidyverse. J Open Source Softw. 2019;4:1686. [Google Scholar]
- 24.Beloeil H, Sulpice L. Peri-operative pain and its consequences. J Visc Surg. 2016;153:S15–8. doi: 10.1016/j.jviscsurg.2016.09.004. [DOI] [PubMed] [Google Scholar]
- 25.Werner MU, Kongsgaard UE. I. Defining persistent post-surgical pain: Is an update required? Br J Anaesth. 2014;113:1–4. doi: 10.1093/bja/aeu012. [DOI] [PubMed] [Google Scholar]
- 26.Jacobsen R, Liubarskiene Z, Møldrup C, Christrup L, Sjøgren P, Samsanaviciene J. Barriers to cancer pain management: A review of empirical research. Medicina (Kaunas) 2009;45:427–33. [PubMed] [Google Scholar]
- 27.Choinière M, Watt-Watson J, Victor JC, Baskett RJ, Bussières JS, Carrier M, et al. Prevalence of and risk factors for persistent postoperative nonanginal pain after cardiac surgery: A 2-year prospective multicentre study. CMAJ. 2014;186:E213–23. doi: 10.1503/cmaj.131012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gunnarsdottir S, Sigurdardottir V, Kloke M, Radbruch L, Sabatowski R, Kaasa S, et al. A multicenter study of attitudinal barriers to cancer pain management. Support Care Cancer. 2017;25:3595–602. doi: 10.1007/s00520-017-3791-8. [DOI] [PubMed] [Google Scholar]
- 29.Al-Atiyyat NM, Vallerand AH. Patient-related attitudinal barriers to cancer pain management among adult Jordanian patients. Eur J Oncol Nurs. 2018;33:56–61. doi: 10.1016/j.ejon.2018.01.009. [DOI] [PubMed] [Google Scholar]
- 30.Can G, Mushani T, Rajhi BH, Brant JM. The global burden of cancer pain. Semin Oncol Nurs. 2019;35:315–21. doi: 10.1016/j.soncn.2019.04.014. [DOI] [PubMed] [Google Scholar]
- 31.Abdel-Razeq H, Mansour A, Jaddan D. Breast cancer care in Jordan. JCO Glob Oncol. 2020;6:260–8. doi: 10.1200/JGO.19.00279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Mills SE, Nicolson KP, Smith BH. Chronic pain: A review of its epidemiology and associated factors in population-based studies. Br J Anaesth. 2019;123:e273–83. doi: 10.1016/j.bja.2019.03.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Peters ML, Sommer M, de Rijke JM, Kessels F, Heineman E, Patijn J, et al. Somatic and psychologic predictors of long-term unfavorable outcome after surgical intervention. Ann Surg. 2007;245:487–94. doi: 10.1097/01.sla.0000245495.79781.65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Bouman EA, Theunissen M, Bons SA, van Mook WN, Gramke HF, van Kleef M, et al. Reduced incidence of chronic postsurgical pain after epidural analgesia for abdominal surgery. Pain Pract. 2014;14:E76–84. doi: 10.1111/papr.12091. [DOI] [PubMed] [Google Scholar]
- 35.Park SK, Yoon S, Kim BR, Choe SH, Bahk JH, Seo JH. Pre-emptive epidural analgesia for acute and chronic post-thoracotomy pain in adults: A systematic review and meta-analysis. Reg Anesth Pain Med. 2020;45:1006–16. doi: 10.1136/rapm-2020-101708. [DOI] [PubMed] [Google Scholar]
- 36.Tait RC, Zoberi K, Ferguson M, Levenhagen K, Luebbert RA, Rowland K, et al. Persistent post-mastectomy pain: Risk factors and current approaches to treatment. J Pain. 2018;19:1367–83. doi: 10.1016/j.jpain.2018.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Nimri O, Arqoub K, Tan’ni N. Jordan: Non-Communicable Diseases Directorate; 2016. Cancer Incidence in Jordan. In: Jordan Cancer Registry, editor. Ministry of Health ed; pp. 1–36. [Google Scholar]
- 38.Clarke H, Katz J, Flor H, Rietschel M, Diehl SR, Seltzer Z. Genetics of chronic post-surgical pain: A crucial step toward personal pain medicine. Can J Anaesth. 2015;62:294–303. doi: 10.1007/s12630-014-0287-6. [DOI] [PubMed] [Google Scholar]
