Decrease of physical fitness during neoadjuvant chemoradiotherapy predicts the risk of pneumonia after esophagectomy

SUMMARY

Although neoadjuvant chemoradiotherapy (nCRT) is frequently used in esophageal cancer patients undergoing treatment with curative intent, it can negatively impact patients’ physical fitness. A decline in physical fitness during chemoradiotherapy may be an indication of vulnerability. The aim of this study was to evaluate whether changes in physical fitness, weight, and fat-free mass index (FFMI) during nCRT can predict the risk of postoperative pneumonia. A retrospective longitudinal observational cohort study was performed in patients who received curative treatment for esophageal cancer between September 2016 and September 2018 in a high-volume center for esophageal cancer surgery. Physical fitness (handgrip strength, leg extension strength, and exercise capacity), weight, and FFMI were measured before and after chemoradiotherapy. To be included in the data analyses, pre- and post-nCRT data had to be available of at least one of the outcome measures. Logistic regression analyses were performed to evaluate the predictive value of changes in physical fitness, weight, and FFMI during nCRT on postoperative pneumonia, as defined by the Uniform Pneumonia Scale. In total, 91 patients were included in the data analyses. Significant associations were found between the changes in handgrip strength (odds ratio [OR] 0.880, 95% confidence interval [CI]: 0.813–0.952) and exercise capacity (OR 0.939, 95%CI: 0.887–0.993) and the occurrence of postoperative pneumonia. All pneumonias occurred in patients with declines in handgrip strength and exercise capacity after nCRT. A decrease of handgrip strength and exercise capacity during nCRT predicts the risk of pneumonia after esophagectomy for cancer. Measuring physical fitness before and after chemoradiotherapy seems an adequate method to identify patients at risk of postoperative pneumonia.

BACKGROUND

Esophageal cancer is the eighth most common cancer worldwide, with an annual incidence of 572 000.¹ Curative treatment involves esophagectomy and is often preceded by neoadjuvant chemoradiotherapy (nCRT) to attain a 5-year survival rate of approximately 40–50%.^2,3 Although nCRT increases overall survival, it can impact a patients’ physical fitness.⁴ Lower levels of physical fitness, defined as muscle strength and exercise capacity, are associated with a lower likelihood of both completing neoadjuvant treatment and undergoing the planned esophagectomy.⁴ Additionally, patients undergoing esophagectomy have a relatively high risk of postoperative complications like anastomotic leakage and pneumonia, and survivors often suffer from decreased physical fitness levels.^5,6 Physical fitness parameters closely interact with nutritional status. Since one out of two patients with esophageal cancer deteriorated in nutritional status during nCRT, both physical fitness parameters and nutritional status like weight and fat-free mass index (FFMI) will be taken into account in this study.⁷ Several systematic reviews show that postoperative pneumonia in cardiac and major abdominal surgery can be reduced by prehabilitation.^8–10 In patients following cardiac surgery, the incidence of postoperative pneumonia even reduced by 50% after a prehabilitation intervention.¹¹ This suggests that postoperative pneumonia might be influenced by the preoperative nutritional state and physical fitness of a patient.

Currently available literature about the predictive properties of physical fitness and postoperative outcomes in patients undergoing esophagectomy seems contradictive.^12,13 Studies measuring physical fitness parameters at only one single preoperative time point are contradictive about the association between preoperative fitness and postoperative outcome.¹⁴ Contrary, in studies that evaluated changes in physical fitness and skeletal muscle mass during nCRT, an association with overall survival after esophagectomy is reported.^4,15,16 Change scores provide insight in patients’ adaptive capacity and resilience to stressors like nCRT. It is suggested that patients with weakened adaptive capacity and resilience to stressors have a greater risk of poor outcomes following surgery.¹⁷ Therefore, change scores of physical fitness, weight, and FFMI during nCRT may be better at predicting postoperative outcomes compared with single measurements. A decline during nCRT may be an indication of vulnerability and therefore imply a higher risk of developing postoperative pneumonia.

Therefore, the aim of this study was to explore the changes in physical fitness, weight, and FFMI during nCRT and to evaluate whether these changes can predict the risk of postoperative pneumonia.

MATERIALS AND METHODS

Study design and participants

A single-center, longitudinal, retrospective observational cohort study was performed at the University Medical Centre (UMC) Utrecht, a high-volume tertiary referral center for esophagogastric surgery. All patients with esophageal cancer undergoing nCRT followed by esophagectomy between September 2016 and September 2018 were eligible to be included. To be included in the data analyses, data from both pre- and post-nCRT visits had to be available of at least one of the outcome measures. The study protocol was assessed and approved by the Medical Ethics Committee of the UMC Utrecht (study protocol number 17/844).

Procedures of usual care

Patients received 5 cycles of nCRT (carboplatin/paclitaxel) and 23 concurrent radiation doses (41.4 Gy) according to the CROSS regimen.² Measurements of physical fitness, weight, and FFMI were performed as part of usual care by an experienced physiotherapist and dietitian before the start of nCRT and 2 weeks after nCRT (Fig. 1). The appointment was part of an outpatient multidisciplinary assessment including the surgeon, nurse specialist, dietitian, and physical therapist. After the nCRT period, which usually takes 5 weeks in total, patients had a planned recovery period of 6–10 weeks until esophagectomy with two-field lymphadenectomy and gastric conduit reconstruction was performed. During the recovery period, patients were advised to be physically active for at least 30 minutes per day. A standardized enhanced recovery protocol was used for perioperative care. This protocol included immediate extubation after surgery and the start of postoperative mobilization on day 1 (Appendix 1).

Outcome measures

Perioperative outcomes

Baseline characteristics included gender, age, body mass index (BMI), nutritional status (Patient-Generated Subjective Global Assessment classification), surgical risk (American Society of Anesthesiologists classification), tumor location, pathological TNM-classification, surgical details (procedure and location of anastomoses), pathological outcomes, length of stay (total and at Intensive Care Unit), and 30-day readmission rate. The primary endpoint of this study was the incidence of postoperative pneumonia, as defined by the uniform pneumonia score.^18,19 Secondary endpoints were the rate of completion of nCRT and the overall rate of postoperative complication with a Clavien–Dindo score of ≥2.²⁰ Baseline and perioperative outcomes were retrieved from a prospectively maintained surgical database.

Physical fitness

Physical fitness measurements included muscle strength (defined as handgrip and leg extension strength) and exercise capacity. Handgrip strength (kg) was evaluated with the Jamar handheld dynamometer (Lafayette Instrument Company, USA). A change of 6 kg is assumed to indicate a clinical relevant difference.^21,22 The Jamar handheld dynamometer has a good to excellent test–retest reproducibility (r > 0.80) and excellent interrater reliability (r = 0.98).²³ Leg extension strength was tested by measuring the quadriceps strength (Newton) with a Hand-Held Dynamometer (Hoggan, microFET 2).²⁴ Exercise capacity (Wpeak) was measured by the SteepRamp Test (SRT). The SRT is a short maximal cycle ergometer test in which patients are asked to cycle at a pedal frequency of 70–80 rpm while the investigator increases the workload by 25 Watt each 10 seconds.²⁵ The test–retest reliability of the SRT is very high (intraclass correlation coefficient of 0.996; 95% confidence interval [CI], 0.989–0.998]) and valid in cancer patients.²⁵

Weight and FFMI

Bioelectrical Impedance Analyses (BIA) measurements were performed using Body Quadscan 4000. Raw BIA data (impedance, resistance, and reactance) at 50 kHz were used to calculate the FFMI, using the Kyle equation.²⁶

Statistical analysis

Physical fitness, weight, and FFMI measurements pre- and post-nCRT were statistically tested by the paired sample t-test. Relative changes in physical fitness, weight, and FFMI during nCRT were calculated in percentages for complete cases; relative change = ([post-nCRT–pre-nCRT)/pre-nCRT] ^*100). To evaluate the predicted value of the relative changes and the risk of pneumonia, multivariate logistic regression analyses were performed separately for each parameter. As age and pulmonary comorbidity are known risk factors for postoperative pneumonia, they were entered in the model as covariates in each multivariable analysis. When a significant association was found with pneumonia, post hoc subgroup analyses were performed. For the post hoc analyses, patients were divided in 4 quartiles based on their relative changes in physical fitness, weight or FFMI. The incidence of pneumonia was presented for each quartile. To gain insight in the potential correlation between the selected parameters, multicollinearity was tested for handgrip strength, leg extension strength, exercise capacity, weight, and FFMI using Pearson correlation coefficients (rating r ≥ 0.50 as moderate and r ≥ 0.80 as strong correlation).²⁷

RESULTS

Patients and surgical characteristics

An overview of the inclusion of patients, and the performed physical fitness, weight, and FFMI measurements is shown in Figure 2. Missing measurements occurred both pre- and post-nCRT. Between September 2016 and September 2018 in total 131 patients underwent esophagectomy with curative intent. Of these in 91 patients there were data pre- and post-nCRT available of at least one of the outcome measures. Since relative change scores could only be calculated in complete cases, only these patients were included in the data analyses.

The mean age was 64 years (standard deviation [SD] 9) and the majority were male (73%). Robot-assisted minimally invasive esophagectomy was most often performed (n = 83, 91%), and almost half of the patients had a cervical anastomosis (n = 38, 42%).

Postoperative outcomes

Overall, of the included 91 patients, 66 patients (73%) developed at least one postoperative complication. Pneumonia occurred in 17 (19%) patients. The median length of stay was 11 days (interquartile range 8–15). All postoperative outcomes are presented in Table 1.

Between patients with a complete case (n = 91) and patients with an incomplete case (n = 40) no statistically significant differences were found in preoperative patient and surgical characteristics. In the postoperative outcomes, significant differences were seen in de incidence of postoperative pneumonia (19% vs. 38%, P = 0.021) and total length of stay (LOS) (median of 11 vs. 15 days, P = 0.040).

Changes and impact of physical fitness, weight, and FFMI

Table 2 shows the level of physical fitness, weight and FFMI of patients before and after nCRT. Significant decreases were seen in exercise capacity (−5%, P = 0.007), weight (−2%, P = 0.000), and FFMI (−2%, P = 0.016). No multicollinearity was found between handgrip strength, leg extension strength, exercise capacity, weight, and FFMI (R ≤ 0.314).

Postoperative pneumonia was significantly more likely to occur in patients who had greater decreases in handgrip strength (odds ration [OR] 0.880 [95%CI 0.813–0.952]) and exercise capacity (OR 0.939 [95%CI: 0.887–0.993]) during nCRT (Table 3). No significant associations were found between changes in leg extension strength, weight, or FFMI and the occurrence of postoperative pneumonia (Table 3). The post hoc analyses showed that none of the patients with stable or improved exercise capacity and handgrip strength developed pneumonia. All pneumonias occurred in patients with declined exercise capacity and handgrip strength (Table 4). No association was found between changes in physical fitness, weight, or FFMI and the risk of overall postoperative complications (Appendix 2).

DISCUSSION

This study describes the changes in physical fitness, weight, and FFMI during nCRT for esophageal cancer and the association of these changes with the risk of pneumonia after esophagectomy. Exercise capacity, weight, and FFMI decreased significantly during nCRT. Besides that, a decrease in handgrip strength and exercise capacity during nCRT is associated with an increased risk of postoperative pneumonia after esophagectomy for cancer. All pneumonias occurred in patients with declined physical fitness during nCRT. Patients who were not included in the data analyses because they missed the physical therapy and/or dietetic counseling pre- and post-nCRT appeared to be the patients with a higher incidence of postoperative pneumonia and prolonged length of stay.

Previous studies demonstrated a significant decrease of exercise capacity during nCRT.^4,28,29 In addition, two previous studies found an association between the change of exercise capacity during nCRT and the risk of postoperative pulmonary complications and overall survival.^4,15 One of these studies showed that patients who improved their physical fitness during nCRT had a lower risk of pulmonary complications compared with patients who remained stable or declined, which is in line with our results.¹⁵ An association was not found between the change in physical fitness during nCRT and overall postoperative complications, which also is in line with a previous study.³⁰ This might be explained by the fact that complications such as anastomotic leakage are also relatively common complications after esophagectomy. Literature showed that there are several numbers of risk factors for the occurrence of anastomotic leakage after esophagectomy.³¹ Therefore, the risk of anastomotic leakage might not be influenced by one factor, such as changes in physical fitness.

This study showed that handgrip strength was associated with the occurrence of postoperative pneumonia, whereas leg extension strength was not. This contrary result might be explained by the fact that the correlation between the two measures was low. Furthermore, literature showed that measuring leg extension strength with a handheld dynamometer is a less reliable instrument compared with the ‘gold standard’ of isokinetic testing.³² However, isokinetic testing is expensive and time consuming. For that reason, a handheld dynamometer was used in the context of usual care in our institute during the study period. To optimize the reliability of measuring leg extension strength, the measurements were performed by a small trained group of physical therapists. Since this study showed that handgrip strength is an adequate method to identify changes during nCRT and can select the patients who are at risk for postoperative pneumonia, this is the recommended method for evaluating muscle strength as predictor for postoperative pneumonia.

In patients undergoing major abdominal surgery, the value of testing preoperative exercise parameters to identify patients at risk for postoperative morbidity is widely studied.^33,34 However, for patients undergoing esophagectomy, literature seems to be contradictive in the relation between physical fitness and postoperative complications, especially when physical fitness measurements were performed at a single moment in time.^{12,13,35–37} Our study showed that relative changes in physical fitness during nCRT can identify patients at risk of postoperative complications. For these patients, exercise training might be beneficial. A previous review showed that preoperative exercise training can result in improved levels of physical fitness in patients with gastrointestinal cancer.³⁸ Therefore, preoperative exercise might be beneficial for patients who are at risk for postoperative pneumonia, particularly patients who show a decline in physical fitness during nCRT. However, further research is needed to explore if a decline in physical fitness is a modifiable risk factor and whether these patients can reduce the risk of the postoperative pneumonia by intensified training exercises before surgery (prehabilitation).

This study has several limitations. First, this study was single centered making the results hard to generalize. Second, missing data occurred. Reasons for missing data were that some patients received their nCRT in another hospital, leading to missing pre-nCRT measurements. Another reason was due to logistic reasons. Patients had a comprehensive program including several disciplines and appointments. Some patients were unable to attend all appointments, missing the physical therapy appointment. Since the missing data were not at random, no imputation was performed. As seen in the results, patients with incomplete data (and therefore not included in the data analyses) had a higher incidence of pneumonia and length of hospital stay after surgery. However, based on the Table 1, these findings cannot be explained by bias in terms of patient and surgical characteristics, since no significant differences were found. Nevertheless, there might have been other factors, not measured in this study, which have played a role in the high incidence of postoperative pneumonia and prolonged length of hospital stay in the missing data. The authors did not have a clear explanation for these differences found in postoperative outcomes. However, it could be that patients who did not complete their preoperative physical therapy and/or dietetic counseling were less well prepared prior to surgery, which might explain a higher incidence of pneumonia and prolonged length of hospital stay.

CONCLUSION

A decrease of handgrip strength and exercise capacity during nCRT predicts the risk of pneumonia after esophagectomy for cancer. Measuring physical fitness before and after chemoradiotherapy might identify patients at risk of unwanted postoperative events and is therefore being suggested as standard of practice. Attention in daily care is needed for the patients who missed the physical therapy and the dietetic counseling in usual care, since these patients have the highest incidence of postoperative pneumonia and prolonged length of hospital stay. Future research should investigate if preoperative physical fitness can be improved in patients awaiting esophagectomy and whether an intensified preoperative exercise program can reduce the risk of complications after esophagectomy for cancer.

Conflicts of interest: The authors declare that they have no conflict of interest.

References

A. Appendix 1

A.1.