The Relationship Between Preoperative PROMIS Scores With Postoperative Improvements in Physical Function After Anterior Cervical Discectomy and Fusion

Article information

Neurospine. 2020;17(2):398-406
Publication date (electronic) : 2020 June 30
doi : https://doi.org/10.14245/ns.1938352.176
Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, IL, USA
Corresponding Author Kern Singh https://orcid.org/0000-0002-6118-7273 Department of Orthopaedic Surgery, Rush University Medical Center, 1611 W. Harrison St, Suite #300, Chicago, IL 60612, USA E-mail: kern.singh@rushortho.com
Received 2019 September 27; Revised 2019 November 13; Accepted 2019 November 23.

Abstract

Objective

Assess preoperative Patient-Reported Outcomes Measurement Information System (PROMIS) physical function (PF) scores and differences between preoperative and postoperative PROMIS-PF scores for patients undergoing anterior cervical discectomy and fusion (ACDF).

Methods

After Institutional Review Board approval, a prospectively maintained surgical registry was retrospectively reviewed for elective spine surgeries of nontraumatic, degenerative pathology between 2015–2018. Inclusion criteria were primary or revision, single-level ACDF procedures. Multilevel procedures and patients without preoperative surveys were excluded. A preoperative PROMIS score cutoff of 35 divided patients into PROMIS-PF score categories (e.g., ≥ 35.0, < 35.0). Categorical and continuous variables were evaluated with chi-square tests and t-tests. Linear regression analyzed PROMIS-PF score improvement.

Results

Eighty-six patients were selected, the high and low PROMIS-PF subgroups only differed in mean age (49.1 vs. 41.3, p = 0.002). Significant differences in PROMIS-PF scores were observed among high and low preoperative PROMIS-PF score subgroups at 6 weeks (p = 0.006), 12 weeks (p = 0.006), and 6 months (p = 0.014). Mean differences between preoperative and postoperative PROMIS-PF scores were significantly different between the high and low PROMIS-PF subgroups at 6 weeks (p = 0.041) and 1 year (p = 0.038). A significant negative association was observed between preoperative PROMIS scores and magnitude of improvement at the 6-week postoperative time point (slope = -0.6291, p < 0.001).

Conclusion

Patients with low preoperative PROMIS-PF scores demonstrated greater improvements at 6 weeks and 1 year. Clinicians should consider patients with low preoperative PROMIS-PF scores to be in the unique position to potentially experience larger postoperative improvement magnitudes than patients with higher preoperative PROMIS-PF scores.

INTRODUCTION

The use of patient-reported outcome measures (PROMs) has increased as a way to assess how patients self-evaluate their own health and capabilities before and after undergoing anterior cervical discectomy and fusion (ACDF) [1-3]. With the advent of newer scoring systems, numerous PROMs have become known as “legacy” measures, with rigorously studied histories of evaluating presurgical and postsurgical outcomes. Several of the “legacy” measures that are most relevant to cervical surgery include questionnaires such as the Neck Disability Index (NDI), the visual analogue scale (VAS) which evaluates levels of pain, and surveys that evaluate overall health, such as the 36-Item ShortForm Health Survey [4]. These “legacy” evaluation systems have demonstrated high levels of reliability and validity [5-8]. Despite these advantages, they can have limited generalizability among various populations and procedure types. Legacy PROMs can also be prone to biases, and are often burdensome for the clinic to administer and for the patient to complete [9-12].

The development of the Patient-Reported Outcomes Measurement Information System (PROMIS) physical function (PF) reflects an improvement in many of these shortcomings, and furthermore, is becoming increasingly validated with respect to ACDF procedures [13,14]. An analysis of PROMIS PF scoring is ideal for ACDF procedures because the system allows for a more global health assessment, fewer questions, and without the need to administer different questionnaires for separate anatomical locations (e.g., arm, neck, etc.). These new strengths increase the utility of PROMIS and may reduce clinical burden.

PROMIS questions are based on a series of domains that include mental, physical, and psychosocial health. Having this broad scope has contributed to an evidence-based reliability that captures relevant decreases in function that might be related to numerous medical conditions. Efficient patient characterizations are achieved by only administering necessary questions for evaluation [6,10]. PROMIS is also unique in that it offers a computer adaptive test (CAT) that customizes questions based on each patient’s response. Research on CAT systems has observed increased evaluation accuracy because the assessments are focused on each patient’s responses, and efficiency is increased because only questions that are necessary for evaluation are asked [9]. While PROMIS, PROMIS PF, and PROMIS PF CAT have different technical referents, in this manuscript, these terms are used interchangeably to refer to the PROMIS PF CAT score. Furthermore, CAT is particularly helpful in evaluating patients with musculoskeletal and spinal pathologies that have undergone surgery. These patients often seek help due to a perceived lack of PF with the goal of improving this dysfunction [9]. Importantly, PROMIS scores can be compared longitudinally, even as the scoring system receives scheduled updates [15].

Though PROMIS PF is most often utilized as a method of evaluation, preoperative scores derived from this metric may be associated with ACDF postsurgical outcomes. The possible influence of preoperative PROMIS PF scores on postoperative PROMIS PF score change has yet to be addressed by current research. The primary aim of this study is to evaluate the association of preoperative PROMIS among patients undergoing ACDF, and to examine mean changes occurring between preoperative and postoperative PROMIS-PF scores. Investigators have previously observed postoperative pain to be associated with the number of vertebral levels on which surgery was performed, irrespective of the spinal region [16]. Hence, this study focused on single-level procedures to limit potential confounding that might be experienced due to variability in PF encountered after operations involving multiple levels.

MATERIALS AND METHODS

1. Patient Population

After receiving Institutional Review Board approval of Rush University Medical Center (ORA #14051301), a prospectively maintained surgical registry of patients undergoing spine surgery between May 2015 and November 2018 was retrospectively reviewed for eligible patients. Inclusion criteria were primary or revision, elective, single-level ACDF procedures. Exclusion criteria included, multilevel procedures and patients without preoperative PROMIS surveys. All patients were treated by a single surgeon at a single academic institution.

2. Data Collection

Baseline patient characteristics and demographics were recorded including, age, sex, smoking status, body mass index (BMI), insurance coverage (Medicare/Medicaid or Workers Compensation/Private), and Charlson Comorbidity Index (CCI). Preoperative comorbid diagnoses were recorded, including acquired immunodeficiency syndrome, arthritis, chronic obstructive pulmonary disease, complicated diabetes, congestive heart failure, gastrointestinal bleeding, hypertension, liver disease, malignancy, metastatic disease, myocardial infarction, neurologic disease, paraplegia, peripheral vascular disease, renal failure, and uncomplicated diabetes. Preoperative spinal diagnoses were recorded, including herniated nucleus pulposus, degenerative disc disease, spinal stenosis, foraminal stenosis, and trauma. Perioperative characteristics were collected, including estimated blood loss (EBL), primary or revision procedure status, operative time (time from skin incision to closure), length of hospital stay, and day of discharge. Fusion rate was evaluated during postoperative clinic visits by either radiograph or computed tomography (CT). Postoperative complications were recorded for all patients. Previously defined cut-points were used to define between “fair” ≥ 35.0 versus “poor” < 35.0 PROMIS score subgroups [17-19]. Patients were categorized by preoperative PROMIS scores (e.g., ≥ 35.0, < 35.0), with higher PROMIS scores representing greater PF. Postoperative PROMIS scores were recorded at the 6-week, 12-week, 6-month, and 1-year time points.

3. Statistical Analysis

Stata 16.0 (StataCorp, College Station, TX, USA) was used to perform a chi-square test analysis to detect for association of PROMIS subgroups in the following demographic variables: sex, smoking status, diagnosis, BMI, and insurance coverage (Table 1). Continuous demographic variables such as age and CCI were assessed with a Student t-test. PROMIS subgroups for operative characteristics including operative time, EBL, fusion rate, length of hospital stay, and day of discharge were also assessed with a Student t-test (Table 2). Postoperative complications were evaluated among subgroups with a chi-square test (Table 3). Mean PROMIS scores at all time points were evaluated using a t-test to detect a difference between PROMIS subgroups (Table 4). Postoperative PROMIS improvement (postoperative–preoperative) was analyzed using a t-test to detect differences at the 6-week, 12-week, 6-month, and 1-year time points (Table 5). Patients who did not fill out a survey at a time point were excluded from the analysis at that time point. A paired t-test compared intragroup pre- and postoperative scores at each time point to evaluate for statistically significant improvement (Table 6). Other PROMs were also assessed in the same manner as PROMIS subgroups, including the VAS arm, VAS neck, NDI and 12-Item Short-Form Health Survey (SF-12) (Table 7). The mean PROMIS score and improvement in PROMIS score were graphed using a scatter plot. Linear regressions were plotted using GraphPad Prism 8.0 for Mac (La Jolla, CA, USA) to evaluate the relationship between preoperative PROMIS scores and postoperative PROMIS score improvements (Figs. 1-5). Statistical significance was set at p < 0.05.

Patient demographics by PROMIS score

Perioperative characteristics by PROMIS score

Postoperative complications

PROMIS score distribution by preoperative subgroup

Postoperative improvement over predetermined time periods

PROMIS score improvement within each subgroup

Patient-reported outcome comparisons by PROMIS physical function group

Fig. 1.

(A) Average PROMIS PF score between ≥ 35 and < 35 preoperative PROMIS PF subgroups. *Significant difference calculated using Student t-test. (B) Postoperative improvement of PROMIS PF score between ≥ 35 and < 35 preoperative PROMIS PF subgroups. *Significant difference calculated using Student t-test. PROMIS PF, Patient-Reported Outcomes Measurement Information System physical function.

Fig. 2.

Linear regression of 6-week difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 3.

Linear regression of 12-week difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 4.

Linear regression of 6-month difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 5.

Linear regression of 1-year difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

RESULTS

1. Demographic Characteristics

A total of 86 patients were identified between May 2015 and November 2018 who underwent primary or revision, single-level ACDF procedures. The cohort was 61.6% male with an average age of 47.3± 9.5 years of which 51.2% were obese (BMI ≥ 30 kg/m2). Table 1 displays baseline characteristics and patient demographics. The 19 subjects in the preoperative PROMIS < 35 group had a significantly younger mean age (41.3 years vs. 49.1 years, p = 0.001) compared to the PROMIS ≥ 35 group. There was no significant difference between groups for the remaining variables including age, smoking status, BMI, insurance, CCI, preoperative comorbid diagnoses, or spinal diagnoses.

2. Perioperative Characteristics

There was no significant difference between groups for any perioperative characteristics including operative time, EBL, hospital length of stay, day of discharge or fusion rate (Table 2). There were no significant differences in the proportion of postoperative complications between groups (Table 3).

3. PROMIS Outcomes

The PROMIS < 35.0 group had significantly lower PROMIS scores preoperatively (29.9 vs. 42.5, p < 0.001), at 6 weeks (35.2 vs 41.7, p = 0.006), at 12 weeks (38.5 vs 47.6, p = 0.006), and at 6 months (39.9 vs 48.3, p = 0.014) postoperatively when compared to PROMIS ≥ 35.0 group (Table 4, Fig. 1A). This difference was lost between subgroups at the 1-year time point (45.9 vs. 48.6, p = 0.591). The PROMIS < 35.0 group had a significantly larger improvement of PROMIS scores when compared to the PROMIS ≥ 35.0 at the 6-week (4.5 vs. -0.84, p = 0.041) and 1-year (15.1 vs. 5.4, p = 0.038) postoperative time points (Table 5, Fig. 1B). At the 12-week and 6-month postoperative time points, the PROMIS < 35.0 group continued to have an increased improvement compared to the PROMIS ≥ 35.0 group, but it was not statistically significant. Both the subgroups demonstrated statistically significant improvements between their pre- and postoperative scores at each time point (Table 6). Following linear regression, a significant negative association was observed between preoperative PROMIS scores and magnitude of improvement at the 6-week postoperative time point (slope=-0.6291, p < 0.001) (Fig. 2). There was no significant association observed at the 12-week (slope=-0.1797, p = 0.3136) (Fig. 3), 6-month (slope=-0.1853, p = 0.3131) (Fig. 4), or 1-year time points (slope=-0.4669, p = 0.0783) (Fig. 5).

4. Other PROM

Statistically significant improvements were observed during the postoperative period for at least one-time point for each score within each subgroup for the 4 other PROMs that were assessed, including VAS arm, VAS neck, NDI, and SF-12 (Table 7). Furthermore, statistically significant differences between each subgroup existed during preoperative evaluations and for multiple postoperative evaluation time points for VAS neck, NDI, and SF-12 (p < 0.05).

DISCUSSION

While cervical surgery outcome evaluations have previously been assessed with “legacy” PROMs such as the SF-12, NDI, and VAS, the next step for increasing accuracy and efficiency in patient assessment is likely to be in the form of CAT metrics such as PROMIS. Though validity of PROMIS in characterizing a patient’s current global health state has been studied and validated by a number of researchers [20-23], others have demonstrated predictive properties of PROMIS [24,25]. In this study, among the 2 samples of individuals having a preoperative PROMIS scores of < 35 or ≥ 35, there were statistically significant differences in the mean improvements at the 6-week and 1-year time period evaluations.

In this assessment, the 2 PROMIS subgroups were divided at a 35.0 cut point based on previously conducted, highly powered studies [19]. Using the 35.0 cut point facilitated the division of 2 subgroups with a statistically significant mean difference in preoperative PROMIS scores. Compared to the PROMIS ≥ 35 group, those with PROMIS scores < 35 had a significantly lower mean preoperative score (42.5 vs. 29.9, p < 0.001) (Table 4). Furthermore, those with lower preoperative PROMIS scores (< 35) were observed to have a larger mean improvement between preoperative PROMIS scores and their evaluations at 6 weeks and 1 year compared to the higher (≥ 35) PROMIS score group. While this difference was statistically significant at the 6-week and 1-year time periods, the comparative mean difference in improvement was not significant between subgroups at 12 weeks and 6 months. Hence, the 2 groups only had a significant difference in PROMIS score improvements at the earliest and latest ends of the evaluation period.

These findings are most relevant in how they might be generalized to support patients with their expectations and experiences during preoperative counseling and throughout recovery. Taking the results of this study, clinicians might be able to identify patients with PROMIS scores of < 35 and reassure them that although their improvement may seem to plateau throughout the mid recovery period, they are likely to still see further recovery toward the end of the recovery period. Likewise, in the patients that began the preoperative period with PROMIS score advantage, it is also important to recognize that such patients are still likely to see improvement throughout the first postoperative year. In the context of this study, both patient groups may see an improvement in global PF as measured by the PROMIS score. Finally, when examining the subgroups by intragroup score, each subgroup was observed to have statistically significant improvements from their mean preoperative score to the mean postoperative score at each timepoint.

This study has several limitations, including the elective surgery patient sample, its retrospective nature, and possible biases in selection and analysis. Evaluation of preoperative PROMIS scores among elective and traumatic patient samples has been observed to affect each subgroup in different ways. An additional limitation that should be considered is that, although fusion rate has been determined, in the majority of cases it was evaluated utilizing X-ray radiographs instead of CT. Confirming boney fusion via radiograph could have led to underreporting of nonunion cases.

As with any retrospective study, selection biases can develop for multiple reasons. Many of the difficulties encountered can be related to possible selection biases. Selection bias can be introduced because of a population that is less generalizable to the greater population. This may occur due to a single institution study design, the investigated population, or confirmation biases among researchers. As with any study focused on outcome measures, including PROMIS, it is often hypothesized that patients who are sicker tend to follow up more frequently than healthy patients. It is difficult to ascertain exactly how loss to follow up would have impacted this patient population since the PROMIS improvement differences were only significantly different at the beginning and end of the observation period. Furthermore, this increasing this study’s observation time would increase the applicability of preoperative PROMIS score and its implications.

A limitation to our study is the assumption that both cohorts of patients will have similar responses to surgical treatment as measured by PROMIS. This assumption may not be true when comparing younger patients who experience acute trauma to older patients with chronic symptoms who are still functionally capable. Further investigation is required to elucidate the temporal effect of symptoms on postoperative PROMIS score improvement. Other limitations may be inherent to the survey. Beyond the requirement for a computer to administer the evaluation, routine behaviors and tasks may not be relatable cross-culturally, and this may hinder the reliability and repeatability of the survey.

Finally, longitudinal evaluations such as PROMIS scoring requires diligent and consistent follow-up. Although the authors exercised their best effort to ensure patients followed up, there was a steady decline in the number of participants that completed each sequential evaluation. As previously stated, this may have occurred due to incentives associated with patients who were sicker (i.e., patients with more severe illness are more eager to follow up). However, one might also consider that sicker patients could have been unable to complete surveys. While there have been a variety of methods established for altering datasets based due to possible effects of selection bias [26], the authors of this study noted the total number of patients at each follow-up (Tables 3-4) and only calculated averages based on the number of observations present at each follow-up. Likewise, changes in scores were not calculated unless the same patient had a score at both evaluated time points.

CONCLUSION

At the early and late evaluations of a 1-year postoperative follow-up, patients with lower preoperative PF scores, as measured by PROMIS, had significantly larger postoperative PROMIS score improvements, i.e., between 6 weeks and 12 months. While lower preoperative PROMIS scores appear to have an inverse relationship with postoperative improvement at the 6-week and 1-year assessments, further observation is required to elucidate if this effect is lessened after the 1-year postoperative time period. Although PROMIS scores can provide a standardized method to evaluate patients before surgery, and throughout their recovery, it also may give clinicians another tool that can be used in the context of a broader clinical picture. Patients with lower preoperative PROMIS scores (< 35) may be counseling regarding the possibly significant changes that are anticipated in their PF during early and late (6 weeks and 1 year) postoperative time periods.

Notes

The authors have nothing to disclose.

References

1. Valderas JM, Alonso J. Patient reported outcome measures: a model-based classification system for research and clinical practice. Qual Life Res 2008;17:1125–35.
2. Andresen AK, Paulsen RT, Busch F, et al. Patient-Reported Outcomes and Patient-Reported Satisfaction after surgical treatment for cervical radiculopathy. Global Spine J 2018;8:703–8.
3. Oliver JD, Goncalves S, Kerezoudis P, et al. Comparison of outcomes for anterior cervical discectomy and fusion with and without anterior plate fixation: a systematic review and meta-analysis. Spine (Phila Pa 1976) 2018;43:E413–22.
4. Khalifeh JM, Dibble CF, Hawasli AH, et al. Patient-Reported Outcomes Measurement Information System physical function and pain interference in spine surgery. J Neurosurg Spine 2019;Apr. 26. :1–10. https://doi.org/10.3171/2019.2.SPINE181237. [Epub].
5. Nwosu K, Schwab J. Patient reported outcome measures for spinal disorders. Semin Spine Surg 2018;30:84–91.
6. McCormick JD, Werner BC, Shimer AL. Patient-reported outcome measures in spine surgery. J Am Acad Orthop Surg 2013;21:99–107.
7. Graves C, Meyer S, Knightly J, et al. Quality in spine surgery. Neurosurgery 2018;82:136–41.
8. Carey TS, Mielenz TJ. Measuring outcomes in back care. Spine (Phila Pa 1976) 2007;32(11 Suppl):S9–14.
9. Brodke DS, Goz V, Voss MW, et al. PROMIS PF CAT outperforms the ODI and SF-36 physical function domain in spine patients. Spine (Phila Pa 1976) 2017;42:921–9.
10. Cella D, Riley W, Stone A, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. J Clin Epidemiol 2010;63:1179–94.
11. Hung M, Cheng C, Hon SD, et al. Challenging the norm: further psychometric investigation of the neck disability index. Spine J 2015;15:2440–5.
12. Hung M, Saltzman CL, Voss MW, et al. Responsiveness of the Patient-Reported Outcomes Measurement Information System (PROMIS), Neck Disability Index (NDI) and Oswestry Disability Index (ODI) instruments in patients with spinal disorders. Spine J 2019;19:34–40.
13. Khechen B, Patel DV, Haws BE, et al. Evaluating the concurrent validity of PROMIS physical function in anterior cervical discectomy and fusion. Clin Spine Surg 2019;32:449–53.
14. Patel DV, Bawa MS, Haws BE, et al. PROMIS physical function for prediction of postoperative pain, narcotics consumption, and patient-reported outcomes following minimally invasive transforaminal lumbar interbody fusion. J Neurosurg Spine 2019;Feb. 1. :1–7. https://doi.org/10.3171/2018.9. SPINE18863. [Epub].
15. Patient-Reported Outcomes Measurement Information System. Physical Function. A brief guide to the PROMIS® Physical Function instruments [Internet]. HealthMeasures; 2019. [cited 2019 Sep 24]. Available from: http://www.healthmeasures.net/images/PROMIS/manuals/PROMIS_Physical_Function_Scoring_Manual.pdf.
16. Bajwa SJ, Haldar R. Pain management following spinal surgeries: an appraisal of the available options. J Craniovertebr Junction Spine 2015;6:105–10.
17. Hays RD, Spritzer KL, Thompson WW, et al. U.S. General population estimate for “excellent” to “poor” self-rated health item. J Gen Intern Med 2015;30:1511–6.
18. Nagaraja V, Mara C, Khanna PP, et al. Establishing clinical severity for PROMIS® measures in adult patients with rheumatic diseases. Qual Life Res 2018;27:755–64.
19. Patient-Reported Outcomes Measurement Information System. Score cut points [Internet]. HealthMeasures; 2019. [cited 2019 Sep 24]. Available from: http://www.healthmeasures.net/score-and-interpret/interpret-scores/promis/promis-score-cut-points.
20. Boody BS, Bhatt S, Mazmudar AS, et al. Validation of Patient-Reported Outcomes Measurement Information System (PROMIS) computerized adaptive tests in cervical spine surgery. J Neurosurg Spine 2018;28:268–79.
21. Brodke DJ, Saltzman CL, Brodke DS. PROMIS for Orthopaedic Outcomes Measurement. J Am Acad Orthop Surg 2016;24:744–9.
22. Haws BE, Khechen B, Bawa MS, et al. The Patient-Reported Outcomes Measurement Information System in spine surgery: a systematic review. J Neurosurg Spine 2019;30:405–13.
23. Purvis TE, Andreou E, Neuman BJ, et al. Concurrent validity and responsiveness of PROMIS health domains among patients presenting for anterior cervical spine surgery. Spine (Phila Pa 1976) 2017;42:E1357–65.
24. Bernstein DN, Houck JR, Gonzalez RM, et al. Preoperative PROMIS scores predict postoperative PROMIS score improvement for patients undergoing hand surgery. Hand (N Y) 2020;15:185–93.
25. Chen RE, Papuga MO, Voloshin I, et al. Preoperative PROMIS scores predict postoperative outcomes after primary ACL reconstruction. Orthop J Sports Med 2018;6:2325967118771286.
26. Howe CJ, Cole SR, Lau B, et al. Selection bias due to loss to follow up in cohort studies. Epidemiology 2016;27:91–7.

Article information Continued

Fig. 1.

(A) Average PROMIS PF score between ≥ 35 and < 35 preoperative PROMIS PF subgroups. *Significant difference calculated using Student t-test. (B) Postoperative improvement of PROMIS PF score between ≥ 35 and < 35 preoperative PROMIS PF subgroups. *Significant difference calculated using Student t-test. PROMIS PF, Patient-Reported Outcomes Measurement Information System physical function.

Fig. 2.

Linear regression of 6-week difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 3.

Linear regression of 12-week difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 4.

Linear regression of 6-month difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Fig. 5.

Linear regression of 1-year difference in postoperative (postop) to preoperative (preop) PROMIS scores in relation to preoperative PROMIS scores. PROMIS, Patient-Reported Outcomes Measurement Information System.

Table 1.

Patient demographics by PROMIS score

Demographic Total (n = 86) PROMIS ≥ 35 (n = 67) PROMIS < 35 (n = 19) p-value
Age (yr) 47.3 ± 9.5 49.1 ± 9.3 41.3 ± 7.3 0.001
Sex 0.361
 Female 33 (38.4) 24 (35.8) 9 (47.4)
 Male 53 (61.6) 43 (64.2) 10 (52.6)
Body mass index 0.506
 Nonobese (< 30 kg/m2) 42 (48.8) 34 (50.8) 8 (42.1)
 Obese (≥ 30 kg/m2) 44 (51.2) 33 (49.3) 11 (57.9)
Smoking status 0.179
 Non-smoker 72 (83.7) 58 (85.6) 14 (73.7)
 Smoker 14 (16.3) 9 (13.4) 5 (26.3)
Insurance coverage (n) 0.592
 Private or WC 85 (98.8) 66 (98.5) 19 (100)
 Medicare/Medicaid 1 (1.2) 1 (1.5) 0 (0)
Ageless CCI 0.7 ± 0.94 0.72 ± 1.0 0.79 ± 0.9 0.767
Preoperative diagnoses*
 Uncomplicated diabetes 10 (11.6) 8 (11.9) 2 (10.5) 0.865
 Congestive heart failure 1 (1.2) 1 (1.5) 0 (0) 0.592
 Hypertension 23 (26.7) 19 (28.4) 4 (21.1) 0.525
 Arthritis 13 (15.1) 9 (13.4) 4 (21.1) 0.413
 Malignancy 1 (1.2) 1 (1.5) 0 (0) 0.592
Spinal pathology
 Herniated nucleus pulposus 73 (84.9) 57 (85.1) 16 (84.2) 0.926
 Degenerative disc disease 7 (8.1) 7 (10.5) 0 (0) 0.142
 Central/spinal stenosis 40 (46.5) 33 (49.3) 7 (36.8) 0.338
 Foraminal stenosis 14 (16.3) 11 (16.4) 3 (15.8) 0.948
 Trauma 0 (0) 0 (0) 0 (0) -

Values are presented as mean±standard deviation or number (%).

PROMIS, Patient-Reported Outcomes Measurement Information System; CCI, Charlson Comorbidity Index; WC, workers compensation.

*

There were no patients in our study with a recorded medical history of myocardial infarction, complicated diabetes, neurologic disease, acquired immunodeficiency syndrome, paraplegia, peripheral vascular disease, metastatic disease, liver disease, renal failure, chronic obstructive pulmonary disease, or gastrointestinal bleeds.

p-value was calculated using Student t-test (continuous), chi-square (categorical), or Fisher exact test (categorical).

Table 2.

Perioperative characteristics by PROMIS score

Characteristic PROMIS ≥ 35 (n = 67) PROMIS < 35 (n = 19) p-value
Procedure status 0.476
 Primary 56 (84.9) 14 (77.8)
 Revision 15 (15.1) 4 (22.2)
Operative time* (min) 57.7 ± 50.8 50.3 ± 10.3 0.531
Estimated blood loss (mL) 28.3 ± 13.7 28.9 ± 9.4 0.856
Length of hospital stay (hr) 10.9 ± 9.0 10.2 ± 8.8 0.747
Day of discharge 0.832
 POD 0 54 (80.6) 15 (79.0)
 POD 1 12 (17.9) 4 (21.0)
 POD 2 1 (1.5) 0 (0)
Fusion rate 64 (98.5) 19 (100) 0.587
 Radiograph 56 (84.9) 12 (63.2)
 CT 10 (15.2) 7 (36.8)

Values are presented as number (%) or mean±standard deviation.

PROMIS, Patient-Reported Outcomes Measurement Information System; POD, postoperative day; CT, computed tomography.

*

Operative time was measured from skin incision to skin closure.

p-value was calculated using Student t-test (continuous), chi-square analysis (categorical), or Fisher exact test (categorical).

Table 3.

Postoperative complications

Postoperative complication PROMIS ≥ 35 (n = 67) PROMIS < 35 (n = 19) p-value
Aspiration 0 (0) 0 (0) -
Urinary retention 0 (0) 0 (0) -
Urinary tract infection 0 (0) 0 (0) -
Acute renal failure 0 (0) 0 (0) -
Postoperative anemia 0 (0) 0 (0) -
Epidural hematoma 0 (0) 0 (0) -
Altered mental status 0 (0) 0 (0) -
Deep venous thrombosis 0 (0) 0 (0) -
Pulmonary embolism 0 (0) 0 (0) -
Pneumothorax 0 (0) 0 (0) -
Arrhythmia 0 (0) 0 (0) -
Ileus 0 (0) 0 (0) -
Pneumonia 0 (0) 0 (0) -
Dysphagia 1 (1.5) 0 (0) 0.592
Atelectasis 0 (0) 0 (0) -
Pleural effusion 0 (0) 0 (0) -
Fever 0 (0) 0 (0) -
Nausea and vomiting 2 (3.0) 1 (5.3) 0.633

Values are presented as number (%).

PROMIS, Patient-Reported Outcomes Measurement Information System.

p-value was calculated using chi-square analysis.

Table 4.

PROMIS score distribution by preoperative subgroup

PROMIS time period PROMIS ≥ 35 PROMIS < 35 p-value
Preoperative 42.5 ± 5.6 (67) 29.9 ± 2.5 (19) < 0.001
6 Weeks 41.7 ± 7.7 (47) 35.2 ± 2.6 (12) 0.006
12 Weeks 47.6 ± 10.0 (38) 38.5 ± 7.9 (12) 0.006
6 Months 48.3 ± 8.8 (37) 39.9 ± 5.7 (8) 0.014
1 Year 48.6 ± 9.0 (27) 45.9 ± 10.2 (4) 0.591

Values are presented as mean±standard deviation (number).

PROMIS, Patient-Reported Outcomes Measurement Information System.

p-value was calculated using Student t-test (continuous).

Table 5.

Postoperative improvement over predetermined time periods

PROMIS time period Postoperative improvement
p-value
PROMIS ≥35 PROMIS <35
Preoperative
 6 Weeks -0.84 ± 8.5 (47) 4.5 ± 3.8 (12) 0.041
 12 Weeks 5.0 ± 8.8 (38) 8.0 ± 7.4 (12) 0.298
 6 Months 6.3 ± 7.9 (37) 9.0 ± 4.7 (8) 0.355
 1 Year 5.4 ± 8.3 (27) 15.1 ± 8.6 (4) 0.038

Values are presented as mean±standard deviation (number).

PROMIS, Patient-Reported Outcomes Measurement Information System.

p-value was calculated using Student t-test (continuous).

Table 6.

PROMIS score improvement within each subgroup

PROMIS time period PROMIS ≥35 p-value PROMIS <35 p-value
Preoperative 42.5 ± 5.6 (67) 29.9 ± 2.5 (19)
6 Weeks 41.7 ± 7.7 (47) 0.500 35.2 ± 2.6 (12) 0.002
12 Weeks 47.6 ± 10.0 (38) 0.001 38.5 ± 7.9 (12) 0.003
6 Months 48.3 ± 8.8 (37) < 0.001 39.9 ± 5.7 (8) 0.001
1 Year 48.6 ± 9.0 (27) 0.002 45.9 ± 10.2 (4) 0.039

Values are presented as mean±standard deviation (number).

PROMIS, Patient-Reported Outcomes Measurement Information System.

p-value was calculated using paired t-test (continuous) to compare each timepoint score to the preoperative value.

Table 7.

Patient-reported outcome comparisons by PROMIS physical function group

Variable PROMIS ≥ 35 p-value* PROMIS < 35 p-value* p-value
VAS arm time period
 Preoperative 5.8 ± 2.4 (62) 6.4 ± 2.7 (16) 0.214
 6 Weeks 2.5 ± 3.2 (62) < 0.001 3.7 ± 4.0 (16) 0.016 0.090
 12 Weeks 2.7 ± 3.3 (53) < 0.001 3.2 ± 4.1 (17) 0.002 0.534
 6 Months 2.7 ± 3.0 (45) < 0.001 3.9 ± 3.6 (13) 0.013 0.162
 1 Year 3.6 ± 3.9 (25) 0.052 3.2 ± 5.0 (5) 0.248 0.863
VAS neck time period
 Preoperative 5.9 ± 1.9 (62) 7.2 ± 2.5 (16) 0.014
 6 Weeks 3.2 ± 2.7 (62) < 0.001 5.3 ± 3.6 (16) 0.048 0.003
 12 Weeks 2.7 ± 2.6 (53) < 0.001 3.9 ± 3.1 (17) < 0.001 0.071
 6 Months 2.5 ± 2.6 (45) < 0.001 3.7 ± 3.5 (14) 0.003 0.002
 1 Year 2.9 ± 2.8 (25) < 0.001 2.8 ± 4.6 (5) 0.147 0.962
NDI time period
 Preoperative 33.6 ± 13.5 (62) 53.6 ± 17.6 (16) < 0.001
 6 Weeks 26.9 ± 18.2 (62) < 0.001 50.6 ± 18.1 (16) 0.509 < 0.001
 12 Weeks 22.4 ± 16.6 (53) < 0.001 40.4 ± 24.2 (17) 0.034 < 0.001
 6 Months 18.2 ± 13.4 (45) < 0.001 40.3 ± 30.7 (13) 0.185 < 0.001
 1 Year 20.8 ± 15.1 (25) 0.004 30.0 ± 41.4 (5) 0.426 0.347
SF-12 time period
 Preoperative 37.3 ± 7.3 (51) 29.2 ± 5.1 (16) < 0.001
 6 Weeks 38.3 ± 9.1 (51) < 0.001 28.2 ± 7.2 (16) < 0.001 < 0.001
 12 Weeks 41.4 ± 9.5 (37) 0.033 31.9 ± 7.9 (15) 0.097 0.001
 6 Months 42.2 ± 9.1 (35) 0.002 34.3 ± 7.9 (11) 0.064 0.019
 1 Year 42.8 ± 6.8 (26) 0.053 32.4 ± 12.1 (5) 0.457 0.055

Values are presented as mean±standard deviation (number).

PROMIS, Patient-Reported Outcomes Measurement Information System; VAS, visual analogue scale; NDI, Neck Disability Index; SF-12, 12-Item Short-Form Health Survey.

*

p-value was calculated using paired t-test (continuous) to compare each timepoint score to the preoperative value.

p-value was calculated using Student t-test (continuous) to compare each timepoint among subgroups.