Bij patiënten met het cervicaal radiculair syndroom (CRS) die een operatie ondergaan is er veel variatie in het advies aangaande de restricties en belastbaarheid van de nek in het postoperatieve beleid. De huidige klinische praktijk is niet gestandaardiseerd ten aanzien van belastbaarheidsadviezen en fysio-/oefenherapeutische nabehandeling. Hierin heerst praktijkvariatie. Ten grondslag aan de praktijkvariatie ligt het feit dat momenteel onduidelijk is welke postoperatieve adviezen rondom fysieke belastbaarheid na de operatie zouden moeten zijn. Deze module evalueert welk postoperatief beleid ten aanzien van belastbaarheid van de nek met meest passend is voor patiënten met CRS.

Conclusions – PICO 1

What are the effects of postoperative advice of activity limitations compared to no physical restrictions in patients who have undergone surgery for CRS, in the acute postoperative phase (first 6 weeks after surgery)?

1a. Quality of life (critical)

1b. Global perceived effect (critical)

1c. Pain (important)

1d. Disability (important)

1e. Return to work (important); 1f. Adjacent level disease (important)

Conclusions – PICO 2

What are the effects of postoperative advice of activity limitations compared to no physical restrictions in patients who have undergone surgery for CRS, after an initial postoperative recovery period (starting 6 weeks after surgery)?

2a. Quality of life (critical)

2b. Global perceived effect (critical)

2c. Pain (important)

2d. Disability (important)

2e. Return to work (important); f. Adjacent level disease (important)

Description of studies - PICO 1 (acute postoperative phase)

Abbott (2013) investigated the physical, functional, and quality of life-related outcomes of patients undergoing anterior cervical discectomy and fusion (ACDF), with and without post-operative activity restrictions (i.e. collar usage). To this end, patients aged 18 to 65 years planned to undergo ACDF were randomized before surgery into the intervention group (n = 17) or the control group (n = 16). During the first days after surgery, both groups received respiratory and circulatory exercises, training of transfers, walking, and activities of daily living by a physiotherapist. Patients in the intervention group received a rigid cervical collar to be worn during daytime over a 6-week period and restrictions from certain activities in the first 3 months after the operation (restricted from activities such as contact sports, running, heavy lifting, driving, and outer-range cervical spine movements). The control group received no postoperative neck movement restrictions. The outcomes quality of life (SF-36), pain (Borg CR-10), and disability (NDI) were assessed after 6 weeks, 3 months, 6 months, 12 months, and 24 months.

Coronado (2020) performed an RCT to examine the acceptability and outcome effects of an early self-directed home exercise program (HEP) within the first 6 weeks after ACDF. Patients aged 21 or older undergoing ACDF were included and randomized to the early HEP (n = 15) or usual care (n = 15). Usual care was administered to both groups and comprised:

• medication,
• cervical collar as indicated (9 patients in HEP-group (60%) and 11 patients in usual care group (73%)),
• driving restrictions (varying from 2 to 6 weeks after surgery), or
• lifting restrictions (not more than 15 pounds or perform sudden or extreme neck movements).

The early HEP was a 6-week self-directed program directly after surgery with walking, sleeping instructions, and range of motion and strengthening exercises performed daily, with personalized adaption by the physiotherapist every 2 weeks. After 6 weeks, 6 months and 12 months, quality of life was measured through the SF-12, arm and neck pain through the NRS, and disability through the NDI.

McFarland (2020) compared clinical outcomes between early cervical spine stabilizer (ECS) training and usual care in patients after ACDF in an RCT. Randomization of patients aged 30 to 75 years and scheduled to undergo ADCF surgery for MRI-confirmed cervical nerve root compression causing radiculopathy took place. Patients were either randomly allocated to ECS training for 6 weeks (n = 20), or usual care for 6 weeks (n = 20). ECS comprised specific instructions with pictures and descriptions of 10 exercises (performed daily with increasing repetitions) for achieving correct positioning and movement; and a walking program. Usual care consisted of a DVD with general spine surgery precautions, and instructions in proper posture, use of cervical collar if applicable, and safety with transfers and walking. Pain (NPRS) and disability (NDI) were the outcomes assessed after 6 and 12 weeks.

Results – PICO 1

1a. Quality of life

Rest versus no rest

Quality of life was reported by Abbott (2013) after 6 weeks, 3 months and 12 months through the SF-36, for patients receiving movement restrictions (secured by prescribing a rigid collar) post-operative compared to patients receiving no advice with regard to activity restrictions.

Exercise versus usual care

Coronado (2020) reported quality of life after 6 weeks and 12 months for patients receiving an early home exercise program and patients receiving usual care, measured through the SF-12. The mean differences in quality of life for the physical (PCS) and mental component score (MCS) between rest and no rest, and exercise and usual care, is depicted in Figure 1a.

For the comparison “rest versus no rest”, rest (movement restrictions and collar) seems to positively affect the PCS (statistically significant), yet possibly negatively affect the MCS; however, this effect on the MCS has disappeared after 12 months. For the comparison “Exercise versus usual care”, a training program does not seem to affect PCS and negatively affect MCS. However, none of the observed differences in PCS and MCS in both comparisons exceed the borders for clinical relevance.

1b. Global perceived effect

No studies reported on the outcome measure global perceived effect.

1c. Pain

Rest versus no rest

Abbott (2013) reported average pain intensity over 24 hours in the neck and shoulder/arm region on the Borg CR-10 scale. This scale was not previously defined (under the heading “Search and select”, subheading “Relevant outcome measures”), yet ranges from 0 (least intense pain) to 10 (most intense pain), and correlates with the VAS (Harms-Ringdahl, 1986). Therefore, the outcomes are reported and depicted in Figure 1b1. Point estimates of pain scores show a clinically relevant effect in favour of movement restrictions, however the confidence interval crosses the border of clinical relevance and significance, implying different inferences.

Exercise versus usual care

Both Coronado (2020) and McFarland (2020) reported pain outcomes through the NRS, after 6 weeks and 12 months, and 6 weeks and 3 months, respectively. Both studies compare an early training programme (home exercise program or cervical spine stabilizer training) to usual care. The results are depicted in Figure 1b2. Observed differences in pain are not statistically significant, nor clinically relevant.

1d. Disability

Rest versus no rest

Abbott (2013) reported disability through the Neck Disability Index (NDI) on a scale of 0 to 50 for patients receiving movement restrictions post-operative compared to patients receiving no advice with regard to activity restrictions. Mean differences are shown in Figure 1d1. Despite being statistically significant at 6 weeks, none of the observed differences in disability are clinically relevant.

Exercise versus usual care

Both Coronado (2020) and McFarland (2020) reported disability outcomes through the NDI: the former on a scale from 0 to 50 after 6 weeks and 12 months; the latter on a scale from 0 to 100 afters 6 weeks and 3 months, respectively. The results are depicted in Figure 1d2. For reading and interpretation purposes the score of McFarland have been divided by 2. No observed differences were statistically significant or clinically relevant.

 1e. Return to work; and

No studies reported on the outcome measure return to work.

1f. Adjacent level disease

Coronado (2020) did not specifically report adjacent level disease but noted that no participants underwent revision surgery during follow-up.

Level of evidence of the literature – PICO 1

1a. Quality of life (critical)

The level of evidence regarding the outcome measure quality of life was downgraded:

6. For Rest versus no rest by 4 levels because of unclear randomization, no blinding, and a large proportion loss to follow-up (-2, risk of bias); and the inclusion of a single study, with confidence intervals of the estimate crossing the border of clinical relevance (-2, imprecision).

7. For Exercise versus usual care by 4 levels because of unclear randomization in one study, insufficient blinding, and selective reporting in one study (-1, risk of bias); not comparing relative rest to no activity restrictions (-1, bias due to indirectness); and the inclusion of a single study, with confidence intervals of the estimate crossing the border of clinical relevance (-2, imprecision).

No downgrading took place for inconsistency or publication bias.

1b. Global perceived effect (critical)

The outcome global perceived effect was not reported and could not be graded.

1c. Pain (important)

The level of evidence regarding the outcome measure pain was downgraded by 3 levels:

8.  For Rest versus no rest because of unclear randomization, no blinding, and a large proportion loss to follow-up (-2, risk of bias); and the inclusion of a single study, with confidence intervals of the estimate crossing the border of clinical relevance (-1, imprecision).
9. For Exercise versus usual care because of unclear randomization in one study, insufficient blinding, and selective reporting in one study (-1, risk of bias); not comparing relative rest to no activity restrictions (-1, bias due to indirectness); the confidence intervals of the clinical estimates crossing the border of clinical relevance (-1, imprecision).

No downgrading took place for inconsistency or publication bias.

1d. Disability (important)

The level of evidence regarding the outcome measure disability was downgraded by 4 levels:

10. For Rest versus no rest because of unclear randomization, no blinding, and a large proportion loss to follow-up (-2, risk of bias); and the inclusion of a single study, with confidence intervals of the estimate crossing the border of clinical relevance (-2, imprecision).
11. For Exercise versus usual care because of unclear randomization in one study, insufficient blinding, and selective reporting in one study (-1, risk of bias); not comparing relive rest to no activity restrictions (-1, bias due to indirectness); and the confidence intervals of the estimates crossing both borders of clinical relevance (-2, imprecision).

No downgrading took place for inconsistency or publication bias.

1e. Return to work (important); and 1f. Adjacent level disease (important)

The outcomes return to work and adjacent level disease were not reported and could not be graded.

Description of studies - PICO 2 (initial postoperative recovery period)

The RCT by Peolsson (2019) investigated the additional benefits of structured postoperative rehabilitation programme (SPT) over a standard approach (SA). Patients with MRI evidence of disc herniation and concomitant clinical signs of cervical radiculopathy for which they underwent surgery, were randomized to either the SPT group (n = 101) or the SA (n = 100) group. Both groups underwent equal postoperative care within the first 6 weeks after surgery. The SPT programme included physiotherapy sessions (from 6 weeks after surgery up until 24 weeks) with individually and progressively adjusted neck-specific exercises and a behavioural approach. The standard postoperative approach comprised usual care with the possibility to seek physiotherapy upon the patient’s own initiative. Outcomes were quality of life, pain, and disability, after 3 months, 6 months, 12 months and 24 months.

Results – PICO 2

2a. Quality of life

Peolsson (2019) reported the quality of life through the EQ-5D questionnaire, scoring from 0 to 1. Mean differences after 3 and 12 months are depicted in Figure 2a (per-protocol analysis). The SMD from ANOVA intention-to-treat analysis between groups was 0.02, indicating no difference in effect between SPT and SA.

Figure 2a. Mean differences in quality of life outcomes between a standardized postoperative rehabilitation program (SPT) and a standard postoperative approach (SA), per-protocol. Higher scores indicate better outcomes.

2b. Global perceived effect

No studies reported on the outcome measure global perceived effect.

2c. Pain

Pain was reported by Peolsson (2019) through the VAS from 0 to 100 mm. The mean difference (from per-protocol data) at 3 months between SPT and SA was 3.0 (95%CI -3.5 to 9.5); and at 12 months 0 (95% CI -7.7 to 7.7). This is not statistically significant nor clinically relevant. The SMD from ANOVA intention-to-treat analysis between groups was 0.07, indicating no difference in effect between SPT and SA.

2d. Disability

The outcome disability was reported after 3 and 12 months through the NDI, on a scale from 0 to 100% (Peolsson, 2019). After 3 months, disability (per-protocol data) was higher in the SPT group (mean difference 4.0; 95% CI -1.0 to 9.0), but after 12 months higher in the SA group (mean difference -2.0; 95% CI -7.8 to 3.8). These differences are not statistically significant, nor clinically relevant. The SMD from ANOVA intention-to-treat analysis between groups was 0.03, indicating no difference in effect between SPT and SA.

2e. Return to work; and 2f. Adjacent level disease

The outcome measures return to work or adjacent level disease were not reported.

Level of evidence of the literature – PICO 2

2a. Quality of life (critical)

The level of evidence regarding the outcome measure quality of life was downgraded by 4 levels because of insufficient blinding, frequent loss to follow-up, and dilution of potential effect due to study methods (-1, risk of bias); not comparing relative rest to no activity restrictions (-1, bias due to indirectness); and the inclusion of a single study, with confidence intervals of the estimate crossing the border of clinical relevance (-2, imprecision). No downgrading took place for inconsistency or publication bias.

2b. Global perceived effect (critical)

The outcome global perceived effect was not reported and could not be graded.

2c. Pain (important)

The level of evidence regarding the outcome measure pain was downgraded by 3 levels because of insufficient blinding, frequent loss to follow-up, and dilution of potential effect due to study methods (-1, risk of bias); not comparing relative rest to no activity restrictions (-1, bias due to indirectness); and the inclusion of a single study that was underpowered for this comparison (-1, imprecision). No downgrading took place for inconsistency or publication bias.

2d. Disability (important)

The level of evidence regarding the outcome measure disability was downgraded by 3 levels because of insufficient blinding, frequent loss to follow-up, and dilution of potential effect due to study methods (-1, risk of bias); not comparing relative rest to no activity restrictions (-1, bias due to indirectness); and the inclusion of a single study that was underpowered for this comparison (-1, imprecision). No downgrading took place for inconsistency or publication bias.

2e. Return to work (important); 2f. Adjacent level disease (important)

The outcomes return to work and adjacent level disease were not reported and could not be graded.

A systematic review of the literature was performed to answer the following questions:

PICO 1: What are the effects of postoperative advice of activity limitations compared to no physical restrictions in patients who have undergone surgery for CRS, in the acute postoperative phase (first 6 weeks after surgery)?

P: Patients who underwent surgery for CRS

I: Advice within first 6 weeks after surgery: activity limitations

C: Advice within first 6 weeks after surgery: no activity limitations/no physical restrictions

O: Quality of life, global perceived effect, pain, disability, return to work, adjacent level disease

PICO 2: What are the effects of postoperative advice of activity limitations compared to no physical restrictions in patients who have undergone surgery for CRS, after an initial postoperative recovery period (starting 6 weeks after surgery)?

P: Patients who underwent surgery for CRS

I: Advice starting 6 weeks after surgery: activity limitations (after initial postoperative recovery)

C: Advice starting 6 weeks after surgery: no activity limitations/no physical restrictions (after initial postoperative recovery)

O: Quality of life, global perceived effect, pain, disability, return to work, adjacent level disease

Relevant outcome measures

The guideline development group considered quality of life and global perceived effect as critical outcome measures for decision making; and pain, disability, return to work and adjacent level disease as important outcome measures for decision making. The working group defined the outcome measures as in Table 1.

The working group defined a 10% difference for continuous outcome measures (weighted mean difference), 20% for dichotomous outcome measures informing on relative risk (RR ≤ 0.91 and ≥ 1.25), and 0.5 for Cohen’s d in standardized mean difference (SMD ≤-0.5 and ≥ 0.5) as minimal clinically (patient) important differences. This decision was based on the minimal important change scores described in the article by Ostelo (2008), in accordance with the Dutch Lumbosacral Radicular syndrome guideline (NVN, 2020).

Table 1. Definitions and instruments of researched outcome measures

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms from 1990 until 17 August 2022. The detailed search strategy is available upon request. The systematic literature search resulted in 382 hits. Studies were selected based on the following criteria:

1. Systematic review and/or meta-analysis, with detailed search strategy, risk of bias assessment, and results of individual studies available; or randomized controlled trial (RCT);
2. Patients aged ≥ 18 years;
3. studies including ≥ 20 (10 in each study arm) patients;
4. studies according to the PICOs (found under the heading “Search and select”);
5. Follow-up period for PICO 1 of 6 weeks, 12 weeks or 3 months, and/or 12 months; and for PICO 2 a period of 3 months/12 weeks and/or 12 months;

• full-text English or Dutch language publication

Thirteen studies were initially selected based on title and abstract screening. After reading the full text, nine studies were excluded (see the table with reasons for exclusion under the heading “Evidence Tables”) and four studies were included (PICO 1 n=3; PICO 2 n=1).

Results

Four studies were included in the analysis of the literature, three RCTs for PICO 1 and one RCT for PICO 2. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.