Enhancing early post-operative physiotherapy input to patients undergoing an oesophagectomy: A quality improvement project

Introduction

Oesophageal cancer is the eighth most common malignancy and the sixth most common cause of cancer-related death worldwide. Surgery is still the only curative therapy option for oesophageal cancer (Tougeron et al. 2011) and long-term survival is poor (Morita et al. 2011). Postoperative pulmonary complications (PPCs) are a serious morbidity following an oesophagectomy with rates as high as 45% (Derogar et al. 2012). Rutegard et al. (2012) carried out a nationwide review of oesophagectomy patients and found that post-operative complications were an independent predictor for poorer long-term survival, even in patients surviving the initial post-operative period. Consequently, strategies to reduce PPCs are of considerable importance (Guinan et al. 2016). Such strategies include the development of enhanced recovery after surgery (ERAS) pathways.

ERAS ensures the patient is given evidence based multidisciplinary care to reduce the risk of complications and help improve outcomes. One element of ERAS is early mobilisation post operatively (NHS Improving Quality 2013). This is not a new concept as the benefits of early mobilisation have been recognised as early as 1946 by Canavarro, who noticed a reduction of PPCs in patients mobilised in the first 24 to 36 hours (Canavarro 1946).

Based on this evidence, and the understanding of the significant risk of complications associated with the oesophagectomy procedure, a local physiotherapy review was completed and identified that despite an ERAS approach, 52% of patients were still being diagnosed with PPCs. It is likely that these PPCs resulted in increased length of stay (LOS), requirement for critical care admissions and increased physiotherapy. All of which required additional healthcare input (and subsequent costs), reduced patient flow and resulted in a worse patient experience.

In response to these complications, a quality improvement (QI) project was initiated to review and improve the model of physiotherapy provided to these patients. The aim of the QI project was to re-prioritise and further enhance physiotherapy input in the early post-operative period, with an anticipated reduction in the incidence of PPCs. Secondary aims were to reduce time to achieve mobility markers and reduce hospital LOS.

Methodology

Design

A QI project utilising the Plan-Do-Study-Act approach. The data collected during the QI project was retrospectively compared to baseline data from the previous year.

Setting

The University Hospital of Wales, part of Cardiff and Vale University Health Board, is a 1000 bed tertiary referral centre for a wide range of clinical services, including neurosciences, transplant and haematology. Additionally, it is the tertiary referral site for those requiring an oesophagectomy from across East and South Wales, with approximately 60 surgical cases completed per annum.

Patients

All patients undergoing an oesophagectomy during the QI intervention period (June– December 2018) were included. These were retrospectively compared to baseline data from patients admitted in a 7-month period during 2017 (April–October).

Quality improvement intervention

The Plan-Do-Study-Act QI programme was utilised to improve early post-operative physiotherapy intervention. A thorough review of the existing physiotherapy model of care was completed. Patients were being seen once a day and then only reviewed in the afternoon if clinically indicated, for example, if they were struggling to clear retained secretions. They were also prioritised the same as all other surgical patients, meaning that they were not guaranteed to be seen due to time constraints and large caseloads. However, it was concluded that due to the high post-operative pulmonary complications (PPCs) occurrence rate and ERAS recommendations, insufficient physiotherapy was being provided during the first 3 post-operative days. As a result, for the QI period, physiotherapy involvement was increased through redistribution of existing resources. The model of care was changed so that all patients undergoing an oesophagectomy were assessed and treated by a physiotherapist twice daily for the first three post-operative days and as clinically indicated thereafter (including weekends where required). No changes were made to pre-operative involvement (nil involvement) or any other aspect of post-operative care.

Data collection

Demographic data was collected for age, gender and surgical type of oesophagectomy. Physiotherapy key performance markers were calculated: time to sit out of bed; time to mobilise independently; length of hospital stay from date of surgery; and occurrence of PPCs. PPCs refer to post-operative pneumonia, atelectasis or other pulmonary complication attributed to undergoing surgery, and were recorded based on Brooks-Brunn classification (Brooks-Brunn 1997). As such, a PPC was confirmed by the presence of two or more of the following, anytime during the first 6 post-operative days:

• New cough/sputum production.
• Abnormal breath sounds compared with baseline.
• Temperature of >38ºC.
• Chest radiograph documentation of atelectasis or new infiltrates.
• Physician documentation of atelectasis or pneumonia.

This information was obtained from medical documentation. Brooks-Brunn classification (1997) was utilised as it has been previously validated for use with patients undergoing abdominal surgery.

Statistics

Data was analysed using SPSS v25 statistical software (SPSS, Chicago, Ill). All statistical tests were 2-sided, and significance was determined at the 0.05 probability level. Simple descriptive data is presented using median and inter-quartile range.

Ethical considerations

This project constituted an improvement in the standard care delivery with no randomisation and thus met the definition of a service evaluation under the NHS Health research authority guidelines. As such ethical approval was not required, and because all outcome measures are collected as part of routine care, the need for consent was waived.

Results

Demographics

All patients requiring an oesophagectomy during the QI period were included for analysis (n = 17). This data was compared to baseline data in 2017 (n = 27). The data is provided in Table 1.

Physiotherapy intervention

During the QI period, the aim was to deliver physiotherapy intervention twice daily during the first three post-operative days. Table 2 provides a percentage overview of the achievement of this target.

Incidence of PPCs

During the QI intervention period the incidence of PPC’s (using Brooks-Brunn classification) was 11 (65%). This compared to an incidence rate of 14 (52%) during the baseline data collection.

Secondary outcomes

Additional data was collected for the achievement of mobility markers, as well as PACU and hospital length of stay (see Table 3).  During the QI period, 3 (18%) patients required re-admission to critical care due to complications (non pulmonary), versus 5 (18.5%) in the baseline period.

Discussion

The aim of this QI project was to improve the physiotherapy model of care for patients undergoing an oesophagectomy and reduce the incidence of PPCs. Comparison of data from baseline, demonstrated that despite increased physiotherapy input there was no reduction in the incidence of PPCs or secondary outcomes.

Demographics

Due to the difference in time of year of data collection, it was not possible to make statistical comparisons in demographic data. The baseline data was collected between April and October 2017, covering a 7-month period. Data in 2018 was collected from June to December. There were also substantially fewer oesophagectomies performed during the QI period (17 versus 27 during baseline).

Despite this, the characteristics of the patient group have remained relatively stable over the last two years. The median age remains 66, most patients were male (89.0% and 88.8% respectively) and a large proportion of these patients were ex-smokers.

However, surgical approach altered between the QI and baseline periods. During the QI data collection, 18% of patients underwent a laparoscopic assisted Ivor-Lewis Oesophagectomy compared to 0% during the baseline. There is a national trend towards increasing the use of laparoscopic approaches with the National Oesophago-gastric Cancer Audit Annual Report (HQIP, 2018) suggesting the proportion of patients who had minimally invasive oesophagectomies was 40.8%. Due to the smaller incision site, and therefore better pain control, these patients may have lower PPC rate.

Physiotherapy intervention

This QI project aimed to provide physiotherapy intervention to patients at least twice daily during the first 3-post-operative days. This was consistently achieved on day one only. By days two and three, only 82% and 18% respectively received twice daily intervention. However, when compared to the baseline data, there was still an improvement for all days except day 3.

The possible reasons for non-completion of the aimed twice daily review were not recorded. However, possible reasons include the reduce physiotherapy service at the weekends and de-prioritisation of patients on day 3 where there was no perceived clinical indication for physiotherapy input. All these patients would have been assessed on an individual basis, and by day 3, could have been independent with sputum clearance techniques, however mobility would often still be limited by attachments. Other possible reasons include return to theatre, completion of scans or return to critical care where pulmonary complication was not the driving factor.

Outcomes

Incidence of PPCs

Despite the increase in physiotherapy intervention, the rate of PPC occurrence increased during the QI project when compared to baseline. Indeed, during the QI period, PPC occurrence was 65%, compared to 52% at baseline despite a higher rate of minimally invasive procedures in the QI period.

An explanation behind the increase in PPCs in 2018 is likely multifactorial. For example, 43% of those patients in the QI project that had PPCs required noradrenaline immediately post operatively. This would have limited how quickly these patients could sit on the edge of the bed or out in the chair, and hence may have increased their risk of PPCs. Common themes associated with PPCs were anastamotic leaks (23.5%) +/- pleural effusions and pneumothoracies (11.7%) (Rutegard et al. 2012).

Of those included in our QI project, one patient had a large pneumothorax and 25% had anastomosis leaks requiring re-admission to critical care, further anaesthetic and surgery, and hence increased likelihood of developing PPCs.

Achievement of mobility markers

Early mobilisation and exercise are known to play an important role in post-operative care and are association with less postoperative reduction of fitness and fewer postoperative complications (Jonsson et al. 2018). As a result, one of the key markers noted in this QI was mobility. The time to first sit on the edge of the bed improved by one day during the QI period, as did time to first sit in the chair. Almeida et al. (2017) state that early mobilisation improves cardiopulmonary endurance, decreases fatigue symptoms, improves muscular strength and quality of life. The time to mobilise with assistance from a physiotherapist was the same during the QI period as the baseline data at 7 days. Mobilising away from the bed space was often limited by attachments.

It is likely that patients are getting out of bed earlier due to the increased early physiotherapy involvement.

Post Anaesthetic Care Unit and hospital length of stay

All patients requiring an oesophagectomy at the host organisation are admitted to the Post Anaesthetic Care Unit (PACU) after surgery. It is planned that after 24–48 hours in PACU the patient then be transferred to the surgical ward.

The median LOS on PACU was 2 days during both the QI and baseline data collection periods. Analysis demonstrated that 44% of patients were discharged from PACU in the first 24 hours during baseline data collection, whereas only 22% of patients were discharged during the first 24 hour period during the QI project. This increase in those staying more than 24-hours is likely a result of a service decision to monitor these patients for longer rather than physiological differences.

A non-statistically significant increase in hospital length of stay (p = 0.888) was observed during the QI period (13.5 days versus 12 days at baseline). This compares to a national average of 11.2 days (National Oesophago-Gastric Cancer Audit 2018). There are numerous possible reasons for the increase in length of stay. For example, higher PPC occurrence rate, an increased number of patients having anastomosis leaks, a reduced number of oesophagectomies being performed, or procedure incision used.

Future service development

Due to the higher PPC rate and the lack of impact increasing physiotherapy input has had on this, further investigation is required on different methods of decreasing the PPC occurrence.

Physiotherapy interventions currently focus on postoperative respiratory function and early mobilisation. These patients all go to PACU post operatively, however they often have issues with pain relief, blood pressure control and fluid administration. All these factors affect physiotherapy input and early mobilisation. Further multi-disciplinary working may identify alternative approaches to ensure appropriate timing of physiotherapy intervention.

Preoperative physiotherapy could also be beneficial. One potential solution to reducing the PPC rate could be seeing patients in a pre-assessment clinic. This would be a useful opportunity to provide education to those patients undergoing an oesophagectomy about physiotherapy involvement, importance of early mobilisation and chest clearance techniques. During a preoperative assessment, it would also be possible to look at risk factors such as co-morbidities and smoking history. Frellick and Barclay (2019) and Boden et al. (2018) discovered that a 30-minute preoperative session halved the number of PPCs and specifically hospital acquired pneumonia in elective patients undergoing upper abdominal surgery.

Depending on a patient’s functional status preoperatively, a prehabilitation service could also be an option. Le Roy et al. (2016) stated that prehabilitation improved postoperative morbidity, length of stay, nutritional and physical status. O’Neil et al. (2017) completed a 12-week multidisciplinary rehabilitation programme consisting of exercise, dietary counselling, and education and found that there was a clinically significant improvement in functional performance and quality of life.

Valkenet et al. (2014) suggest that using inspiratory muscle training preoperatively would prevent the imbalance between ventilation demand and ventilation capacity post oesophagectomy, and therefore reduce the PPC rate.

Limitations

One of the major limitations of this QI project was the sample size. This was smaller than the baseline data due to fewer surgeries being completed during the data collection period. The lower sample size makes it more difficult to identify significant relationships from the data (Hackshaw 2008). Future QI projects might consider data collection over a longer time period to increase the sample size.

Another limitation was the use of the Brooks-Brunn classification tool for PPCs. This was used in the QI for continuity and a more accurate comparison of the two sets of data. However, this classification is not overly sensitive to PPCs that physiotherapy can impact. A large proportion of patients will have sputum production postoperatively following the anaesthetic, and a raised temperature could be due to several things, for example, an anastomosis leak, infection or other medical complications. In the future it would be worth looking at other PPC classification tools to ensure that they are more sensitive to PPCs that physiotherapy can impact.

Conclusion

In this QI project, enhancing physiotherapy input in the first 3 post-operative days had no effect on reducing the incidence of PPCs. Similarly, there was no change in the achievement of mobility markers or hospital length of stay. Further work is now required to explore other interventions such as prehabilitation and their impact on PPC occurrence.

de Almeida, E., Almeida, J., Landoni, G., Galas, F., Fukushima, J., Fominskiy, E., de Brito, C., Cavichio, L., de Almeida, L., Ribeiro, U., Osawa, E., Pe Diz, P., Cecatto, R., Battistella, L., Hajjar, L. (2017). Early mobilization programme improves functional capacity after major abdominal cancer surgery: A randomized controlled trial. BMJ: British Journal of Anaesthesia 119(5), 900–907. 

Atkins, Z. and D’Amico, T. (2006). Respiratory complications after esophagectomy. Thoracic Surgery Clinics 16(1), 35–48. 

Boden, I., Skinner, E., Browning, L., Reeve, J., Anderson, L., Hill, C., Robertson, I., Story, D., Denehy, L. (2018). Preoperative physiotherapy for the prevention of respiratory complications after upper abdominal surgery: Pragmatic, double blinded, multicentre randomised controlled trial. British Medical Journal 360: j5916

Brooks-Brunn, J. (1997). Predictors of post-operative pulmonary complications following abdominal surgery. Chest 111(3), 564–571. 

Canavarro, K. (1947). Early postoperative ambulation. Annals of Surgery 124(2), 180–181. 

Derogar, M., Orsini, N., Sadr-Azodi, O., Lagergren, P. (2012). Influence of major postoperative complications on health-related quality of life among long-term survivors of esophageal cancer surgery. Journal of Clinical Oncology 30(14), 1615–1619.

Doorakkers, E., Konings, P., Mattsson, F., Lagergren, J., Brusselaers, N. (2015). Early complications following oesophagectomy for cancer in relation to long-term healthcare utilisation: A prospective population-based cohort study. PLOS ONE 10(3).

Frellick, M. and Barclay, L. (2019). Can preop physiotherapy prevent postop respiratory complications? Medscape https://www.medscape.org/viewarticle/893646.

Guinan, E., Dowds , J., Donohoe, C., Reynolds, J., and Hussey, J. (2017). The physiotherapist and the esophageal cancer patient: From prehabilitation to rehabilitation. Disease of the Esophagus 30(1), 1–12. 

Hackshaw, A. (2008). Small studies: Strengths and limitations. European Respiratory Journal 32(5), 1141–114. 

Health and Social Care Information Centre. (2018). National Oesophago-Gastric Cancer Audit 2018. 

Huang, L. and Onaitis, M. (2014). Minimally invasive and robotic Ivor Lewis esophagectomy. Journal of Thoracic Disease 6(3), 314–321. 

Koppert, L., Lemmens, V., Coebergh, J., Steyerberg, E., Wijnhoven, B., Tilanus, H. and Janssen-Heijnen, M. (2012). Impact of age and co-morbidity on surgical resection rate and survival in patients with oesophageal and gastric cancer. British Journal of Surgery 99(12), 1693–1700. 

Jonsson, A., Lin, E., Patel, L., Patel, A., Stetler, J., Prayor-Patterson, H., Singh, A., Srinivasan, J., Sweeney, J. and Davis, S. (2018). Barriers to enhanced recovery after surgery after laparoscopic sleeve gastrectomy. Journal of the American College of Surgeons 226(4), 605–613. 

Roy, B., Pereira, B., Bouteloup, C., Costes, F., Richard, R., Selvy, M., Pétorin, C., Gagnière, J., Futier, E., Slim, K., Meunier, B., Mabrut, J., Mariette, C. and Pezet, D. (2016). Effect of prehabilitation in gastro-oesophageal adenocarcinoma: Study protocol of a multicentric, randomised, control trial – the PREHAB study. British Medical Journal  6(12):e012876

NHS Improving Quality. (2013). Enhanced Recovery after Surgery Guideline. Available at: http://www.1000livesplus.wales.nhs.uk/publications.

Morita, M., Nakanoko, T., Kubo, N., Fujinaka, Y., Ikeda, K., Egashira, A., Saeki, H., Uchiyama, H., Ohga, T., Kakeji, Y., Shirabe, K., Ikeda, T., Tsujitani, S. and Maehara, Y. (2011). In-hospi- tal mortality after a surgical resection for esophageal cancer: Analyses of the associated factors and historical changes. Annals of Surgical Oncology 18(6), 1757–1765. 

O’Neill, L., E. Guinan, E., Doyle, S., Elliott, J., O’Sullivan, J., Reynolds, J. and Hussey, J. (2017). Rehabilitation strategies following esophageal cancer (the ReStOre trial): A feasibility study. Diseases of the esophagus 30(5), 1–8. 

Rutegård, M., Lagergren, P., Rouvelas, I., Mason, R. and Lagergren, J. (2012). Surgical com- plications and long-term survival after esophagectomy for cancer in a nationwide Swedish cohort study. European Journal of Surgical Oncology 38(7), 555–561. 

Tougeron, D., Richer, J. and Silvain, C. (2011). Management of esophageal adenocarcinoma. Journal of Visceral Surgery 148(3), 161–170. 

Valkenet, K., Trappenburg, J., Gosselink, R., Sosef, M., Willms, J., Rosman, C., Pieters, H., Scheepers, J., Heus, S., Reynolds, J., Guinan, E., Ruurda, J., Rodrigo, E., Nafteux, P., Fon- taine, M., Kouwenhoven, E., Kerkemeyer, M., van der Peet, D., Hania, S., Hillegersberg, R. and Backx, F. (2014). Preoperative inspiratory muscle training to prevent postoperative pulmonary complications in patients undergoing esophageal resection (PREPARE Study): Study Protocol for a Randomized Controlled Trial. Trials 15, 1, 144.

Weledji, E. and Verla, V. (2016). Failure to rescue patients from early critical complications of oesophagogastric cancer surgery. Annals of Surgery and Medicine 7(10), 34–41.