“In the tube” following sternotomy: A quasi-experimental study

Abstract

Background

Traditionally, physical movement has been limited for cardiac surgery patients, up to 12-weeks post-operatively. Patients are asked to use “standard sternal precautions,” restricting their arm movement, and thereby limiting stress on the healing sternum.

Aim

To compare return to function, pain/discomfort, wound healing, use of pain medication and antibiotics, and post-operative length of hospital stay in cardiac surgery patients having median sternotomy who used standard sternal precautions or Keep Your Move in the Tube movement protocols post-operatively.

Methods

A quasi-experimental design was used (100 standard sternal precautions and 100 Keep Your Move in the Tube patients). Patients were followed in person or by telephone over a period of 12-weeks postoperatively. Outcomes were measured at day 7, as well as weeks 4, 8, and 12 weeks.

Results

The majority of participants (77% in each group) were male and had coronary artery bypass graft surgery (66% standard sternal precautions and 72% Keep Your Move in the Tube). Univariate analysis revealed the standard sternal precautions group had lesser ability to return to functional activities than the Keep Your Move in the Tube group (p < 0.0001) over time. This difference was minimized however, by week 12. Multivariate analysis revealed that increasing age, body mass index, and female sex were associated with greater functional impairment over time, but no difference between standard sternal precautions and Keep Your Move in the Tube groups.

Conclusions

Keep Your Move in the Tube, a novel patient-oriented movement protocol, has potential for cardiac surgery patients to be more confident and comfortable in their recovery.

Introduction

Despite spiraling use of percutaneous coronary interventions, coronary artery bypass graft (CABG) as well as valve surgery remain common approaches for the treatment of coronary artery and valve diseases.¹ Estimates suggest that over one million open heart surgeries are performed annually worldwide.² The majority of open heart surgeries require a median sternotomy to gain access to the heart. Healing of a median sternotomy is an arduous process encompassing both physical (e.g. pain, discomfort, wound healing, functional impairment) and psycho-social components (e.g. quality of life).^2,3

Patients are usually advised, following sternotomy, to limit specific activities to enable time for the sternum to heal sufficiently and to have confidence that there will be no adverse outcome associated with activity. Often labelled as “sternal precautions,” patients are cautioned against activities such as “lifting, pushing, or pulling” weights of greater than 5–10 pounds, driving, or using their arms to assist with sitting or rising.⁴ Patients are also advised to splint their chests (wrapping their arms around their chests) when moving or coughing. “Sternal precautions” are recommended for a period of up to 12 weeks post-operatively: until sternal union is fully achieved. Some have argued that sternal precautions may be too restrictive.⁵ Further, these precautions can have detrimental effects of their own, including muscle atrophy^4,6 and inability to resume activities of daily living. The procedures for use of sternal precautions have no universal definition,⁷ can vary widely,^7,8 and there is no empirical evidence to support their use.⁵

Keep Your Move in the Tube (KYMITT^TM) is a post-sternotomy movement protocol in which the aim is to keep the patients’ arms from being outstretched, limiting movement of the humerus, and thereby reduce potential tension on the sternotomy.⁴ This protocol is based on ergonomics and patient education (not direction). “By keeping their upper arms close to their body, as if they were inside an imaginary truncal tube, patients can modify load-bearing movements and thus avoid excessive stress to the sternum” (p.98).⁴ Patients are taught how to get out of bed, get out of a chair, etc. by physiotherapists and nurses who use established graphics to reinforce the “tube” concept. The key difference of using this approach relative to traditional sternal precautions is that no time limit is imposed on patients to return to comfortable physical activity. Patients are instructed to “stay in the tube” until they are able to perform movements without pain or discomfort. Then, they are instructed to gradually move out of the “tube.” This is an innovative, practical and patient-oriented approach that our team evaluated in a single center in western Canada.

Objective

Our objective was to compare return to function, pain/discomfort, wound healing, use of pain medication, and antibiotics, and post-operative length of hospital stay in cardiac surgery patients having median sternotomy who used standard sternal precautions (SSP) or KYMITT^TM movement protocols post-operatively. The study protocol was approved by the Conjoint Health Research Ethics Board at the University of Calgary and was conducted in accordance with the World Medical Association Declaration of Helsinki.

Methods

Design

A quasi-experimental design was used. Diffusion of the intervention, meaning that patients randomized to the non-intervention arm would be exposed to the intervention by watching patients randomized to the intervention arm thus creating a threat to internal validity, would make a randomized controlled trial impossible. The first group of sternotomy patients received instructions in keeping with SSP. Following a period of 8 weeks to provide extensive training to physiotherapy and nursing staff regarding the KYMITT^TM movement protocol, and to allow time for all SSP patients to be discharged from hospital, the second group of post-operative patients received KYMITT^TM instructions. All patients receiving either SSP or KYMITT^TM instructions were cautioned to immediately restrict their activities should they notice sternal movement (clicking) and to report this to the investigators and their surgeon.

Participants

Though all patients admitted for cardiac surgery through median sternotomy received either SSP or KYMITT^TM instructions, only patients who met the inclusion/exclusion criteria were asked to participate in this study. Consecutive patients who had cardiac surgery (e.g. CABG, valve, congenital repair) through median sternotomy during the recruitment periods, and who were aged 18 years or older were invited to participate in the study. Patients who did not speak or read English, presented with cognitive impairment (e.g. confusion, dementia, Alzheimer’s disease, current substance abuse) that would preclude completion of survey questions, or did not have telephone access, were not asked to participate in the study. Written informed consent was received from all study participants. Patient recruitment began in October 2018 and continued through to May 2019.

Sample size calculation

The sample size calculation was based on data from a pilot trial in which sternal closure was examined.⁹ The calculation was based on the scores from the measure of functional status (Health Assessment Questionnaire (HAQ)¹⁰ disability index). The scores range from 0–3, with zero being not disabled and three being very disabled. We used the mean scores from day 7, as well as weeks 4, 8, and 12. The minimally important clinical difference of mean scores is 0.22.¹¹ Comparing (https://www.stat.ubc.ca/~rollin/) two independent samples, two-sided test, average standard deviation 0.5,^9,11 alpha of 0.5, and desired power 0.80, a sample size of 82 for each group was required. Allowing for attrition, 100 patients were recruited for each group.

Outcomes

The primary outcome was return to function over time (specifically at day 7, and weeks 4, 8, and 12). Other outcomes included: pain, wound healing, pain medication use, and antibiotic use (measured at day 7, and weeks 4, 8, and 12), as well as post-operative length of stay. It was unnecessary to include sternal wound infection (superficial or deep), sternal dehiscence, or mediastinitis as outcomes as they occur in only 0.2–5% of sternotomy patients.^1,12 Time until moving out of the “tube” was examined in KYMITT^TM participants. Patients who had previously undergone sternotomy were asked to make comment about the differences in their experiences related to use of SSP or KYMITT^TM. Processes and outcomes regarding patient safety were monitored closely throughout the study period.

Procedure

Potential participants were approached to consent on the third day post-operatively. Data were collected in person while the patient was hospitalized. Over that time, they were taught how to assess their sternal wound, which was successfully done by patients in the Women’s Recovery from Sternotomy Trial (WREST) study, with 97% agreement.¹³ Other relevant clinical data (e.g. age, grafts used for and nature of surgery, analgesic use, antibiotic use for sternal wound) were collected through a health record audit at the time of discharge.

Post-discharge follow-up began on their next scheduled visit. Pain, discomfort, wound healing, as well as return to function data using the HAQ¹⁰ were collected at the first home telephone contact. At all contacts, participants were asked to, on a 24-hour recall basis, indicate their daily medication use (analgesic and antibiotic – prescribed for sternal symptoms only). Participants in the KYMITT^TM group were also asked if they had “left the tube.” At the end of the study, KYMITT^TM group patients who had undergone a previous sternotomy were asked to comment about their experiences with each movement protocol.

Functional status

Function was measured using the HAQ¹⁰ to capture participants’ difficulty in performing activities of daily living. The HAQ¹¹ was originally developed for assessment of patients with rheumatoid arthritis, but has been used with a wide range of patients, including cardiac surgery patients.^9,13,14 The participant is asked to reflect over the past week and identify how much difficulty he/she has had (without any, with some, with much, unable to do) managing a variety of activities of daily living encompassed in eight dimensions (dressing and grooming, arising, eating, walking, hygiene, reach, grip, activities). The appeal of using this questionnaire is its focus on activities in which patients wish to engage during the early part of their sternotomy recovery – focusing on both fine and gross motor skills affected by sternotomy-related pain and discomfort. In addition, the HAQ¹⁰ may be used confidently in interview situations, without visual assessment of the patient and takes less than 10 min to complete. A disability index is calculated (0–3 (mild–severe)).

Pain

Pain was measured using an 11-point numeric rating (Likert-type) scale, anchored with descriptors 0=none, 10=worst. Using a numeric rating scale to measure pain (though not rendering ratio level data) has been shown to be a very quick, sensitive, reliable, and valid means of measurement.^15,16

Wound healing

Wound healing was classified using relevant descriptors for drainage, stability and incision approximation, on a five-point ordinal scale. These descriptors are consistent with those used in clinical practice and the Society of Thoracic Surgeons database. They were successfully used in the WREST study¹³ where participants were easily taught to assess the status of their own incisions.

Analgesic and antibiotic use

Analgesic (as an indicator of pain/discomfort) and antibiotic use (as an indicator of sternal infections) data were collected by health record audit (at hospital discharge) and by 24-hour recall interview at telephone follow-up. These data were reported as yes/no.

Post-operative length of stay

A health record audit was undertaken at hospital discharge. Operative date and date of discharge were noted, and post-operative length of stay was calculated.

Leaving the tube

KYMITT^TM participants were asked if and when they “left the tube” at each telephone follow-up visit. Their responses were recorded as “yes” or “no.”

Comparison of experiences

Patients who had undergone a previous sternotomy were asked to make comment regarding the SSP versus KYMITT^TM experiences. Their responses were summarized by the research assistants.

Data management

Entry into the study and data collection were managed using a secure website (Research Electronic Data Capture) managed through the Epidemiology Coordinating and Research Centre, University of Alberta (which hosts the Consultation and Research Services Platform of the Alberta Strategy for Patient-Oriented Research Support Unit).

Data analysis

Analysis was performed by using R 3.4.0 (Vienna, Austria; https://www.R-project.org/) and SAS 9.4 software (SAS Institute Inc. Cary, North Carolina, USA). Descriptive statistics were used to compare demographic and clinical data by SSP and KYMITT^TM groups and by time points. Univariable analysis was performed using Chi-square test or Fisher’s test (when small frequencies were present) for categorical variables, and t-test or Wilcoxon rank sum test (when data were heavily skewed) for continuous variables to test the association/difference between groups and time points. The wound healing scores were skewed such that we dichotomized the scores (as done in the WREST study)¹⁴ with <3 no problems and ≥⩾3 having wound healing problems. An unbalanced analysis of variance (ANOVA) for two-way design with interaction was used to check if the means of HAQ scores between time points, groups, and interaction between them were significantly different by generalized linear modeling (GLM). Regression analysis was performed to quantify the effect of time points and group on HAQ¹⁰ score by presenting the coefficient (standard error), 95% confidence interval and p-value. Variables entered into the multi-variate analysis were chosen based on the univariate analysis findings and known factors associated with sternal wound healing. The performance of model was measured by R² and adjusted R². A p-value <0.05 was considered statistically significant.

Safety

A strict safety protocol was in place. Research assistants for this study had collected outcomes data for previous sternal wound healing studies^9,13 and were instructed should high scores be recorded for indicators (e.g. wound healing, pain), that the patient should be referred to their family physician or surgeon for urgent assessment and a follow-up call made the following day to ensure that this had occurred.

A data monitoring committee (three members, including a biostatistician) was established to undertake an interim analysis. These were clinical researchers external to the study, who reviewed all outcomes data when 10 patients in the KYMITT^TM group completed 6 weeks of post-operative follow-up. This analysis was completed, and the study proceeded.

Results

As seen in Figure 1, 250 patients were screened to enter this study and 200 patients (100 in SSP and 100 in KYMITT^TM) signed consents. Patient data were included in the analysis if data were available for three of the five timepoints. This rendered a total of 166 patients (81 in SSP and 85 in KYMITT^TM).

The majority of patients were male (64.45%), with significantly more males in the KYMITT^TM group than the SSP group (p = 0.029). The mean age of the patients was 66.10 years (standard deviation (SD)=9.05) and the mean body mass index (BMI) was 29.41 kg/m² (SD=5.02). Though the KYMITT^TM group appeared to have significantly fewer staples and more sutures used for sternal wound skin closure, there were missing data. Otherwise, there were no differences in mean age or BMI between groups (see Table 1).

Overall, the patients in the KYMITT^TM group reported significantly lower HAQ scores (p < 0.0001), indicating greater return to function than those in the SSP group (see Figure 2). There was some attenuation of the rate of return to function in the KYMITT^TM group relative to the SSP group between weeks 8 and 12, thus minimizing the overall difference by week 12. This is possibly due to an interaction with reduction in pain over time in both groups (see the following).

As expected, and as seen in Table 2, there was a statistically significant improvement in pain scores over time for both groups (p < 0.001). However, there were no significant differences in pain scores between groups over time (p = 0.529).

There were no statistically significant differences in wound healing problems (p = 0.135), though KYMITT^TM patients had fewer wound healing problems than SSP patients overall (see Table 3). There were no significant differences between groups in use of pain medications (p = 0.426) or antibiotics (p = 0.682) over the course of the study. Finally, there were no significant differences between groups in length of hospital stay (p = 0.239).

Sex, age, BMI, and previous sternotomy were incorporated as covariates into a mixed effects model to examine return to function over time (see Table 4). As found in the univariate analyses, there was a statistically significant average change in the HAQ score for each subsequent trial week (week 4, 8, and 12) compared to day 7, with p-values <0.001, but no difference between SSP and KYMITT^TM groups. For each unit increase in age and BMI, there was a small increase (meaning reduced function) in the HAQ score. Though statistically significant, this was not a clinically meaningful finding given that the minimally important clinical difference of mean scores is 0.22.¹¹ Female sex was associated with a statistically significant and clinically meaningful reduction in function. It appears the HAQ score was higher (meaning greater functional impairment) for patients without previous sternotomy relative to those who had previous sternotomy. The R² (variance explained by fixed factors) for this model was 68.28% and the adjusted R² (variance explained by fixed and random factors) was 83.85%.

Two thirds of the KYMITT^TM patients had left “the tube” by the 12-week follow-up (Table 5). Those patients who had not completely left “the tube” indicated they continued to use the tube from time to time during their day and some called it their “safe place.”

Seven of the KYMITT^TM participants had undergone previous sternotomy. All were asked to make a comparison regarding their experience with KYMITT^TM relative to their experience with SSP (see Table 6). Three of the seven indicated they could not recall much difference and were more neutral about using KYMITT^TM. However, the other four participants were much more positive.

Discussion

The key finding from this study is that patients using the KYMITT^TM movement protocol may enable return to function more quickly in the early post-operative period, and at no greater risk for adverse outcomes (i.e. pain, wound healing problems, antibiotic use, or post-operative length of stay) than those patients using SSP. We believe this early functional improvement will be meaningful to patients. Some patients who had experience using both movement protocols made comment regarding the relatively positive experience using KYMITT^TM movement protocol. Importantly, we did not find any evidence of harm as a result of this change in practice.

The KYMITT^TM movement protocol is patient-oriented, meaning that patients make the decisions regarding how and when to extend their movements “out of the tube” based on their subjective experience of pain or discomfort. Brocki et al.⁵ identified “if sternal precautions are logical and practicable in daily activity, then the patient is more likely to act accordingly, not out of fear of damaging one self, but because it makes sense" (p.78).⁵ One patient identified this approach as a “do-system” rather than a “don’t do-system.” Indeed, this is a “patient supportive approach focusing on possibilities and not restrictions" (p.82).⁵

Increase in age and BMI, as well as being female, were associated with reduced functional improvement. These findings are consistent with those from other studies. There is evidence to suggest that obese cardiac surgery patients have a slower return to function following cardiac surgery relative to those who are leaner, although this phenomenon is also associated with older age.¹⁷ Females are also known to have a slower return to function than their male counterparts.^18,19

We did not ask caregivers (e.g. physiotherapists, nurses, or family members) regarding their experiences using the KYMITT^TM movement protocol relative to SSP. Caregivers of cardiac surgery patients are faced with many demands, including assisting with mobilization. “Lifting or transferring dependent patients" (p.349)²⁰ is a major risk factor for reporting back injuries for physiotherapists,²⁰ nurses,²¹ as well as family members.^22,23 Given that the KYMITT^TM movement protocol enables cardiac surgery patients to have greater ability and flexibility in using their hands to aid in moving out of bed or rising from a chair relative to SSP, it would be reasonable to assume that there will be less potential for caregivers to be overburdened with physically assisting the patient. Future research regarding use of the KYMITT^TM movement protocol may include assessing healthcare provider injury rates following its implementation.

There were certain limitations to this study. As indicated earlier, it was not reasonable to use a randomized controlled trial design in this single center study. The authors acknowledge that a larger scale, multi-center, randomized controlled trial would have been the gold standard to enhance both internal and external validity. There were some potentially important differences between groups (i.e. sex and wound closure) that may have been eliminated had the participants been randomized to treatment groups. There was also a >15% loss of patients for the analysis, given that we included only those who had data for three of the five timepoints of interest. This was a greater loss than our previous post-cardiac surgery study (WREST study).¹³ Finally, the data collected for this study were primarily subjective, leaving open the possibility of bias.

Conclusion

KYMITT^TM is a patient-oriented and “sensible” movement protocol, based on sound ergonomics and patient education.⁴ Given our findings regarding earlier return to function in those patients using the KYMITT^TM movement protocol, there is potential for cardiac surgery patients to be more confident and comfortable in their recovery without any adverse effects. There is also potential for reduction in healthcare provider and family caregiver burden that should be explored.

Acknowledgements

The authors acknowledge the support of the Consultation and Research Services Platform at The Alberta SPOR SUPPORT Unit in statistical services. The authors wish to thank the Libin Cardiovascular Institute of Alberta for their “in kind” contributions. The authors acknowledge the immense work of Pamela LeBlanc, in overseeing the study implementation as well as the assistance of Fern Marusiak, Krystyna Mitchell, and Rebecca Carriere, who assisted with patient recruitment and data collection. Finally, the authors acknowledge the enthusiastic support and encouragement of Jenny Adams, Richard Gach, and Susan Triano who not only guided KYMITT^TM’s implementation but laid the foundation for success.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded, in part, by the College and Association and the University of Alberta’s Department of Physical Therapy (Clinical Research Fund).

References