Summary of findings:

Prehabilitation (nutrition, physical, or multimodal) compared to no prehabilitation (standard care, no prehabilitation, placebo) for patients awaiting colorectal surgery

Patient or population: patients awaiting colorectal surgery

Intervention (I): prehabilitation (nutrition, physical, or multimodal)

Comparison (C): no prehabilitation (standard care, no prehabilitation, placebo)

Outcomes

Results

№ of
studies

Certainty of the evidence
(GRADE)

Conclusion

Mortality (nutritional interventions)

Four RCTs described mortality (Braga 2002; Burden 2017; MacFie 2000; Sorensen 2014). No single RCT found a significant difference between treatment and control groups.

Braga (2002) reported 1 death in one of the intervention groups receiving immunonutrition (1/50 and 0/50) and 1 death in one of the control groups (1/50 and 0/50). Burden (2017) also observed 1 death in the intervention group while there were 4 deaths in the control group (1/55 versus 4/45, p>0.05). One death in both the intervention groups (1/24 and 1/24) was observed by MacFie (2000) while there were 2 and 1 deaths in the control groups (2/27 and 1/25). Sorensen (2014) reported 2 death in both the intervention and control group (2/74 versus 2/74).

(4 RCTs)

⨁⨁◯◯
LOW ^a

Nutritional prehabilition might not increase or decrease mortality.

Mortality (physical exercise interventions) - not reported

-

-

The effect of physical prehabilitation on mortality is unknown.

Mortality (multimodal interventions) - not reported

-

-

The effect of multimodal prehabilitation on mortality is unknown.

Morbidity, adverse events, and complications (nutritional interventions)

Infectious complications:

Only two of the six RCTs (Braga 2002; Burden 2011; Burden 2017; Finco 2007; MacFie 2000; Sorensen 2014) found significant differencs between groups. Braga (2002) reported that participants receiving preoperative immunonutrition had significant less infectious complications than participants who did not receive immunonutrition (5/50 and 5/50 versus 16/50 and 15/50, p<0.02 and p<0.04). Burden (2017) found a significant difference in surgical site infections between groups (intervention: 11/55, control: p=0.044), where the intervention group had 11 surgical site infections (11/55) compared to 17 surgical site infections in the control group (17/45). There were no differences for chest infections (5/55 versus 3/45, p=0.359) or urinary tract infections (4/55 versus 6/45, p=0.315) reported in Burden (2017).

Overall complications:

Overall complications were reported in six RCTs (Burden 2011; Burden 2017; Gilis 2016; MacFie 2000; Smedley 2004; Sorensen 2014). None of these six trials found significant differences between groups.

Adverse events:

Four RCTs reported adverse events (Braga 2002; Finco 2007; MacFie 2000; Sorensen 2014). Reported events were postoperative cramps and bloating, nausia, abdominal discomfort, diarrhea. When groups were compared, only Braga (2002) found a significant difference between groups for postoperative cramps/bloating (prioperative supplements: 14/50, preoperative supplements: 6/50, conventional care: 5/50, no supplements: 7/50, p<0.001). Burden (2017) reported that 7 participants found themselves intolerant for the intervention by reporting symptoms of nausia (n=4), abdominal discomfort (n=3), and/or diarrhea (n=2).No single adverse event was obsered by Sorensen (2014) related to the intake of supplements in the intervention or control group.

(8 RCTs)

⨁⨁◯◯
LOW ^b,d

Nutritional prehabilitation might not increase or decrease (infectious) complications or adverse events.

Morbidity, adverse events, and complications (physical exercise interventions)

Overall complications:

Carli (2010) observed 22/56 participants having postoperative complications in the interventiongroup, while 18/54 participants had complications in the control group (p-value not reported). Dronkers (2010) found no significant difference between intervention and control group (9/22 versus 8/20, p=0.65). Moug (2019) observed in both the intervention and control group 12 participants with complications (12/18 versus 12/22).

Adverse events:

Dronkers (2010) and Moug (2019) reported that there were no (serious) adverse events. Kim (2009) had two participants dropping out in the intervention group due to feelings of fatigue and malaise.

(4 RCTs)

⨁⨁◯◯
LOW ^b,c

Physical exercise prehabilitation might not increase or decrease complications or adverse events.

Morbidity, adverse events, and complications (multimodal interventions)

Overall complications:

Number of participants with at least one complication were reported in two RCTs (Bousquets-Dion 2018; Gilis 2014). Both RCTs did not find significant differences between groups (14/37 versus 8/26, p=0.562, and; 12/38 versus 17/39, p=0.277).

(2 RCTs)

⨁◯◯◯
VERY LOW ^a,b

Multimodal prehabilitation might not increase or decrease complications.

Hospital stay (nutritional interventions)

Eight RCTs described the length of hospital stay (Braga 2002; Burden 2011; Burden 2017; Finco 2007; Gilis 2016; MacFie 2000; Smedley 2004; Sorensen 2014). Only one RCT (Braga 2002) found a significant difference between groups, where the intervention groups had a significant shorter mean hospital stay than the control groups (9.8 and 9.5 days versus 12 and 12.2 days). In one RCT it was not clear whether a statistical test was performed, however the median length of stay was 13.5 days for the intervention group versus 14 days for the control group. Six other RCTs did not find significant differences between groups for the length of hospital stay.

(8 RCTs)

⨁⨁◯◯
LOW ^b,c

Nutritional prehabilitation might not increase or decrease the length of stay

Hospital stay (physical exercise interventions interventions)

Three RCTs reported the length of stay after surgery (Carli 2010, Dronkers 2010; Moug 2019). Carlie found a mean stay of 11.9 days (SD: 36.6) in the intervention group and a mean stay of 6,6 (SD: 3.6) in the control group (p-value not reported). Dronkers (2010) found a shorter mean stay in the intervention group (16.2 days (SD: 11.5)) than in the control group (21.6 days (SD: 23.7)), however this difference was not statistically significant (p=0.31). Mough (2019) reported a median stay of 11 days (range: 6-37) for the intervention group, while the controlgroup had a median stay of 10 days (range: 0-38).

(3 RCTs)

⨁◯◯◯
VERY LOW ^b,c

We are unsure of the effect of physical exercise prehabilitation on the length of stay

Hospital stay (multimodal interventions)

Two RCTs described the length of stay (Bousquets-Dion 2018; Gilis 2014). Bousquets-Dion (2018) stated that the length of stay was comparable in both groups (no data or p-value reported) and Gilis (2014) found a median stay of 4 days (IQR: 3-5) for the intervention group and 4 days (IQR: 3-7) for the control group (p=0.812).

(2 RCTs)

⨁◯◯◯
VERY LOW ^b,c

We are unsure of the effect of multimodal prehabilitation on the length of stay

Care (in)dependency (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on care (in)dependency is unknown

Care (in)dependency (physical exercise interventions) - not reported

-

-

-

-

The effect of physical prehabilitation on care (in)dependency is unknown

Care (in)dependency (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on care (in)dependency is unknown

Quality of life (nutritional interventions)

Gilis 2016 reported the mental and physical component of the SF-36 questionnaire, however found no significant difference. Smedley 2004 reported that there were no significant differences between groups with the SF-36 and EuroQol questionnaires.

(2 RCTs)

⨁⨁◯◯
LOW ^b,c,e

Nutritional prehavilitation might not increase or decrease the quality of life

Quality of life (physical exercise interventions)

Dronkers (2010) did not find any significant differences between groups on the EORTC-QLQ C30 global health scale (p=0.88), functional scale (p=0.72), and symptom scale (p=0.20). Moug (2019) only reported the fatigue subscale of the EORTC-QLQ C30. One to two weeks before surgery, the intervention group scored a mean of 28.8 points (SD: 23.9, changescore: 0.7 (95%CI: -13.7 to 15) from baseline) on the subscale, while the control group scored a mean of 25.6 points (SD: 16.9, change score: 7.2 (95%CI: 1.8 to 12.6 from baseline)).

(2 RCTs)

⨁◯◯◯
VERY LOW ^b,c

We are unsure of the effect of physical exercise prehabilitation on the quality of life

Quality of life (multimodal interventions)

One trial reported results for quality of life (Gilis 2014). No significant differences were found between groups for any of the SF-36 subscales in the period from baseline to 8 weeks after surgery: physical functioning (p=0.468), role physical (p=0.360, bodily pain (p=0.623), general health (p=0.98), vitality (p=0.597), social functioning (p=0.656), role emotional (p=0.672), and mental health (p=0.085).

(2 RCTs)

⨁◯◯◯
VERY LOW ^b,c,f

We are unsure of the effect of multimodal prehabilitation on the length of stay

Coping (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on coping is unknown

Coping (physical exercise interventions) - not reported

-

-

-

-

The effect of physical prehabilitation on coping is unknown

Coping (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on coping is unknown

Participation (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on participation is unknown

Participation (physical exercise interventions) - not reported

-

-

-

-

The effect of physical prehabilitation on participation is unknown

Participation (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on participation is unknown

Compliance (nutritional interventions)

Three trials did not find statistically significant differences between groups (Braga 2002; Gilis 2016; Sorensen 2014). Braga (2002) compared the mean intake between groups (890ml, 905ml, 915ml). Gilis (2016) compared the mean compliance to the precription (93.7% versus 96.6%, p=0.51). Sorensen (2014) reported the number of compliant participants until surgery (64/74 versus 59/74, p=0.381) and after surgery (17/74 versus 16/74, p=1.00), however significant less participants were compliant after surgery than before surgery (p=0.001).

Burden (2011) reported the number of participants compliant to the intervention (100% compliant: 36/54, 75% compliant: 8/54, <25% compliant: 6/54). Burden (2017) reported the compliance to the intervention in a similar manner (100% compliant: 29/39, 75% compliant: 2/39, 50% compliant: 3/39, 25% compliant: 3/59, 0% compliant: 2/39).

(5 RCTs)

⨁⨁⨁◯
MODERATE ^c

There is probably a fair amount of compliance to nutritional interventions.

Compliance (physical exercise interventions)

Dronkers (2010) reported an attendance of 97% for the physical exercise training. Kim (2009) found that participants exercised on 74% (SD: 16, range: 44-95%) of the available training days.

(2 RCTs)

⨁◯◯◯
VERY LOW ^b,c

We are unsure of the compliance to physical exercise prehabilitation.

Compliance (multimodal interventions)

Bousquet-Dion (2018) did not find any significant differences for the compliance to physical exercise at any time point:

preoperative:

98% (intervention only)

4 weeks postoperatively:

72% (intervention)

79% (control)

8 weeks postoperatively:

82% (intervention)

75% (control)

In the compliance to dietry advice in Bousquet-Dion (2018) no significant differences were found either:

preoperative:

100% (intervention only)

4 weeks postoperatively:

91% (intervention)

84% (control)

8 weeks postoperatively:

92% (intervention)

83% (control)

Gilis (2014) reported 78% (SD: 21) compliance preoperatively in the intervention group. Four weeks after surgery a significant difference in compliance was found, where the intervention group was 53% (SD: 30) compliant compared to 31% (SD: 26) in the control group (p<0.001). In the period between 4 and 8 weeks after surgeryno significant difference was found: 53% compliance (SD: 33) in the intervention group and 40% (SD: 31%) in the control group (p=0.117).

(2 RCTs)

⨁◯◯◯
VERY LOW ^b,c

We are unsure of the compliance to multimodal interventions

Patient satisfaction (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on patient satisfaction is unknown

Patient satisfaction (physical exercise interventions)

Participants in Moug (2019) were asked about their satisfaction with the trial on a 5-point likert scale (0: completely disagree, 5: completely agree). Participants in the intervention group scored their satisfaction with the trial as a median of 5 (IQR: 5-5), while the control group scored a median of 4.5 (IQR: 4-5).

(1 RCT)

⨁◯◯◯
VERY LOW ^c,g

We are unsure of the effects of physical exercise on the patient satisfaction.

Patient satisfaction (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on patient satisfaction is unknown

Patient burden (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on patient burden is unknown

Patient burden (physical exercise interventions) - not reported

-

-

-

-

The effect of physical prehabilitation on patient burden is unknown

Patient burden (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on patient burden is unknown

Patient investment (nutritional interventions) - not reported

-

-

-

-

The effect of nutritional prehabilitation on patient investment is unknown

Patient investment (physical exercise interventions) - not reported

-

-

-

-

The effect of physical prehabilitation on patient investment is unknown

Patient investment (multimodal interventions) - not reported

-

-

-

-

The effect of multimodal prehabilitation on patient investment is unknown

Costeffectiveness (nutritional interventions)

-

-

-

The costeffectiveness of nutritional prehabilitation is unknown

Costeffectiveness (physical exercise interventions) - not reported

-

-

-

-

The costeffectiveness of physical prehabilitation is unknown

Costeffectiveness (multimodal interventions) - not reported

-

-

-

-

The costeffectiveness of multimodal prehabilitation is unknown

*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

CI: Confidence interval

GRADE Working Group grades of evidence

High certainty: We are very confident that the true effect lies close to that of the estimate of the effect
Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different
Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect
Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Auteur

Interventie (n)

Controle (n)

Type interventie

Braga 2002

I₁: perioperatieve orale suppletie met arginine en omega-3 vetzuren (preoperatief: drank, postoperatief: enteraal) (n=50)

I₂: preoperatieve orale suppletie met arginine en omega-3 vetzuren (n=50)

C₁: preoperatieve suppletie met een isonitrogeneus en isoenergetisch vloeibaar dieet. (n=50)

C₂: conventionele zorg zonder kunstmatig dieet voor of na chirurgie. (n=50)

Nutritie

Bousquet-Dion 2018

I: Voedingsadviezen en angstreductie. Preoperatief lichamelijke oefeningen die thuis worden uitgevoerd, en eenmaal wekelijk een oefensessie onder supervisie. Postoperatief lichamelijke oefeningen in het ziekenhuis (3 dagen postoperatief) met voortzetting van de oefeningen in de thuisomgeving tot en met 8 weken postoperatief. (n=41)

C: Voedingsadviezen en angstreductie.

Postoperatief lichamelijke oefeningen in het ziekenhuis (3 dagen postoperatief) met voortzetting van de oefeningen in de thuisomgeving tot en met 8 weken postoperatief. (n=39)

Multimodaal

Burden 2011

Preoperatief orale supplementen (proteïne) en dieetadviezen (n=54)

Dieet adviezen (n=62)

Nutritie

Burden 2017

I: Preoperatief orale supplementen (proteïne) en schriftelijke dieetadviezen (n=55)

C: Preoperatief water drinken en schriftelijke dieetadviezen (n=46)

Nutritie

Carli 2010

Preoperatief fietsoefeningen en weerstandstraining (n=58)

Preoperatief wandeloefeningen en ademhalingsoefeningen (n=54)

Fysiek

Dronkers 2010

Weerstandstraining, aerobe training, en ademhalingsspier training (n=22)

Adviezen over thuis oefenen (n=20)

Fysiek

Finco 2007

Preoperatieve voeding met arginine, omega-3 vetzuren, RNA) én postoperatief enterale immunonutritie (n=14)

Dieet met een laag vezel gehalte (n=14)

Nutritie

Gilis 2014

Preoperatief én postoperatief: gematigde aerobe oefeningen, weerstandstraining, voedingsadviezen met proteïne supplementen, ontspanningsoefeningen (n=38)

Alleen postoperatief: gematigde aerobe oefeningen, weerstandstraining, voedingsadviezen met proteïne supplementen, ontspanningsoefeningen (n=39)

Multimodaal

Gilis 2016

Geïndividualiseerde voedingsadviezen en perioperatief wei proteïne supplementen (n=22)

Geïndividualiseerde voedingsadviezen (n=21)

Nutritie

Kim 2009

Preoperatieve aerobe training (n=14)

Ademhalingsoefeningen en cardio oefeningen (n=7)

Fysiek

MacFie 2000

I₁: preoperatieve én postoperatieve supplementen (proteïne en koolhydraten) (n=24)

I₂: preoperatieve supplementen (proteïne en koolhydraten) (n=24)

C₁: alléén postoperatieve supplementen (proteïne en koolhydraten) (n=27)

C₂: Geen supplementen (n=24)

Nutritie

Moug 2019

I: preoperatieve loop oefeningen (n=24)

C: standaard zorg met advies om het normale niveau van fysieke activiteit te behouden. (n=24)

Fysiek

Smedley 2004

I₁: preoperative én postoperative supplementen (proteïne) (n=32)

I₂: preoperatieve supplementen (proteïne) (n=41)

C₁: postoperatieve supplementen (proteïne) (n=35)

C₂: Geen supplementen (n=44)

Nutritie

Sorensen 2014

Preoperatieve én postoperatieve orale supplementen verrijkt met omega-3 vetzuren (n=74)

Preoperatieve én postoperatieve orale supplementen (zonder omega-3) (n=74)

Nutritie

Uitkomstmaat

Auteur

Duur follow-up

Resultaat

Mortaliteit

Braga 2002

Tot en met 8 dagen postoperatief

Mortaliteit, n:

I₁: 1/50

I₂: 0/50

C₁: 0/50

C₂: 1/50

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd).

Burden 2017

Tot 30 dagen na chirurgie

Mortaliteit, n:

I: 1/55

C: 4/45

Geen significant verschil tussen groepen (p>0.05)

MacFie 2000

Onduidelijk, mogelijk tot 6 maanden na ontslag.

Mortaliteit, n:

I₁: 1/24

I₂: 1/24

C₁: 2/27

C₂: 1/25

Vergelijkbaar voor alle groepen (onduidelijk of hier een statistische toets voor gedaan is, p-waarde niet gerapporteerd)

Sorensen 2014

Onduidelijk, mogelijk tot 30 dagen na operatie

Mortaliteit, n:

I: 2/74

C: 2/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Morbiditeit, adverse events, en complicaties

Braga 2002

Tot en met 8 dagen postoperatief

Aantal infectieuze complicaties, n:

I₁: 5/50

I₂: 6/50

C₁: 16/50

C₂: 15/50

Significante verschillen voor de groepen die immunonutritie ontvingen ten opzichte van de controlegroepen (p<0.02 voor perioperatieve interventie versus controle en conventioneel; p<0.04 voor preoperatief versus controle en conventioneel)

Aantal non-infectieuze complicaties, n:

I₁: 5/50

I₂: 4/50

C₁: 3/50

C₂: 4/50

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd).

Postoperatieve krampen/opgeblazen gevoel, n:

I₁: 14/50

I₂: 6/50

C₁: 5/50

C₂: 7/50

Significant verschil tussen groepen (p<0.001).

Diarree, n:

I₁: 3/50

I₂: 2/50

C₁: 4/50

C₂: 4/50

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Drie deelnemers moesten postoperatief overgeven waarbij er geen verschil tussen groepen gevonden werd (groepsallocatie niet weergegeven, geen p-waarde gerapporteerd).

Burden 2011

Onduidelijk (postoperatief uit het medisch dossier)

Totaal aantal met een infectie (volgens de definitie van Center for Disease Control), n:

I: 20/54

C: 27/62

Geen significant verschil tussen groepen (p=0.271)

Totaal aantal met een infectie (volgens de definitie van Buzby), n:

I: 20/54

C: 20/62

Geen significant verschil tussen groepen (p=0.587)

Totaal aantal met een (infectieuze) complicatie (volgens de definitie van Buzby), n:

I: 24/54

C: 26/62

Geen significant verschil tussen groepen (p=0.785)

Burden 2017

Tot 30 dagen na chirurgie

Aantal deelnemers met een complicatie, n:

I: 23/55

C: 25/45

Geen significant verschil tussen groepen (p=0.114)

Wondinfecties, n:

I: 11/55

C 17/45

Significant verschil tussen groepen (p=0.044)

Thorax infecties, n:

I: 5/55

C: 3/45

Geen significant verschil tussen groepen (p=0.359)

Urinebuis infecties, n:

I: 4/55

C: 6/45

Geen significant verschil tussen groepen (p=0.315)

Zeven deelnemers in de interventiegroep rapporteerden een intolerantie voor het toegediende supplement. Symptomen waren misselijkheid (n=4), abdominaal discomfort (n=3) en/of diarree (n=2)

Finco 2007

Onduidelijk, mogelijk tot 1 maand na chirurgie

Totaal aantal infecties, n:

I: 3/14

C: 3/14

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Postoperatieve misselijkheid door enterale voeding, n:

I: 2/14

C: 2/14

Geen significant verschil (p-waarde niet gerapporteerd)

Postoperatieve diarree (>3 afscheidingen per dag) door enterale voeding, n:

I: 0/14

C: 0/14

Geen significant verschil (p-waarde niet gerapporteerd)

Postoperatief overgeven door enterale voeding, n:

I: 2/14

C: 1/14

Geen significant verschil (p-waarde niet gerapporteerd)

Gilis 2016

Tot 30 dagen na chirurgie

Ten minste één complicatie, n:

I: 8/22

C: 9/21

Geen significant verschil tussen groepen (p=0.75)

MacFie 2000

Onduidelijk, mogelijk tot 6 maanden na ontslag.

Totaal aantal complicaties, n:

I₁: 6/24

I₂: 7/24

C₁: 6/27

C₂: 3/25

Vergelijkbaar voor alle groepen (onduidelijk of hier een statistische toets voor gedaan is, p-waarde niet gerapporteerd)

Septische complicaties, n:

I₁: 5/24

I₂: 6/24

C₁: 4/27

C₂: 2/25

Vergelijkbaar voor alle groepen (onduidelijk of hier een statistische toets voor gedaan is, p-waarde niet gerapporteerd)

Postoperatieve misselijkheid bij een gemiddelde postoperatieve voedingsduur van 8 dagen (range: 0-20), n

I₁: 5/24

I₂: 0/24

C₁: 3/27

C₂: 0/25

Geen statistische toets uitgevoerd.

Smedley 2004

Onduidelijk

Totaal aantal complicaties, n:

I₁: 15/32

I₂: 20/41

C₁: 15/35

C₂: 34/44

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Aantal kleine complicaties, n:

I₁: 10/32

I₂: 17/41

C₁: 13/35

C₂: 30/44

Significant verschil tussen interventie 1 versus controle 1 en controle 2 (p<0.05). Significant verschil tussen interventie 2 versus controle 1 en controle 2 (p<0.05).

Aantal grote complicaties, n:

I₁: 5/32

I₂: 3/41

C₁: 2/35

C₂: 4/44

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddeld aantal complicaties per patiënt:

I₁: 0,31

I₂: 0,41

C₁: 0,37

C₂: 0,68

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Sorensen 2014

Tot 30 dagen na de operatie

Algehele infectie complicaties, n:

I: 28/74

C: 27/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Algehele non-infectie complicaties, n:

I: 25/74

C: 19/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Ten minste één complicatie, n:

I: 38/74

C: 39/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Deelnemers rapporteerden geen adverse events door de inname van supplementen in zowel de interventie- als controlegroep.

Opnameduur

Braga 2002

Tot ontslag

Gemiddelde opnameduur, dagen (SD):

I₁: 9,8 (3,1)

I₂: 9,5 (2,9)

C₁: 12 (4,5)

C₂: 12,2 (3,9)

Significant verschil tussen interventie 1 versus controle 1 en controle 2 (p<0.0001). Significant verschil tussen interventie 2 versus controle 1 en controle 2 (p<0.0005).

Burden 2011

Tot ontslag

Mediane opnameduur, dagen:

I: 13,5

C: 14

Onduidelijk of er een statistische toets werd uitgevoerd (p-waarde niet gerapporteerd)

Burden 2017

Tot ontslag

Mediane opnameduur, dagen (IQR):

I: 7 (4-10,5)

C: 7 (4-10)

Geen significant verschil tussen groepen (p=0.63)

Finco 2007

Tot ontslag

Gemiddelde opnameduur, dagen (SD):

I: 7,7 (2,3)

C: 6,8 (1,6)

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gilis 2016

Tot ontslag

Mediane opnameduur, dagen (IQR):

I: 5 (3-13)

C: 4 (3-10)

Geen significant verschil tussen groepen (p=0.78)

MacFie 2000

Tot ontslag

Gemiddelde opnameduur, dagen:

I₁: 11

I₂: 12

C₁: 10

C₂: 13

Vergelijkbaar voor alle groepen (onduidelijk of hier een statistische toets voor gedaan is, p-waarde niet gerapporteerd)

Smedley 2004

Tot ontslag

Gemiddelde opnameduur, dagen (SD)

I₁: 11,7 (5,1)

I₂: 12,8 (4,5)

C₁: 13,4 (7,5)

C₂: 14,1 (6,6)

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd).

Sorensen 2014

Tot ontslag

Opnameduur ≤ 5 dagen, n:

I: 14/74

C: 10/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Opnameduur ≥ 10 dagen, n:

I: 35/74

C: 40/74

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd).

Kwaliteit van Leven

Gilis 2016

Vier weken na chirurgie

SF-36 fysieke component, mediaan (IQR):

I: 41,3 (34,2-46,5)

C: 36,5 (34,5-42,8)

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

SF-36 mentaal component, mediaan (IQR):

I: 47,7 (38,1-53,8)

C: 41,3 (35,6-55,8)

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Smedley 2004

4 weken na ontslag

Geen verschillen tussen groepen in de kwaliteit van leven, gemeten met de SF-36 en EuroQol (data en p-waarden niet gerapporteerd).

Compliance

Braga 2002

Tot chirurgie

Gemiddelde inname van het orale supplement, ml:

I₁: 890

I₂: 905

C₁: 915

C₂: niet relevant

Geen significante verschillen tussen groepen (p-waarde niet gerapporteerd)

Burden 2011

Tot chirurgie

Inname van het supplement in de interventiegroep, n:

100% voltooid: 36/54

50% voltooid: 8/54

<25% voltooid: 6/54

Vier personen hadden het innamepatroon niet vastgelegd.

Burden 2017

Tot chirurgie

Compliance aan interventie o.b.v. de hoeveelheid ingenomen supplementen, n:

I: 29/39 waren 100% compliant; 2/39 waren 75% compliant; 3/39 waren 50% compliant; 3/39 waren 25% compliant; 2/39 waren 0% compliant

Gilis 2016

Onduidelijk, mogelijk tot 4 weken na chirurgie

Compliance, % (95% BHI):

I: 93,7% (86-100%)

C: 96,6% (91-100%)

Geen significant verschil tussen groepen (p=0.51)

Sorensen 2014

Tot chirurgie en onduidelijk: na chirurgie

Compliance tot chirurgie, n compliant:

I: 64/74

C: 59/74

Geen significant verschil tussen groepen (p=0.381)

Compliance na chirurgie, n compliant:

I: 17/74

C: 16/74

Geen significant verschil tussen groepen (p=1.00). Significant minder deelnemers compliant dan tot chirurgie (p=0.001)

Uitkomstmaat

Auteur

Duur follow-up

Resultaat

Morbiditeit, adverse events, en complicaties

Carli 2010

Onduidelijk

Postoperatieve complicaties, n:

I: 22/56

C: 18/54

Onduidelijk of er een statistische toets werd uitgevoerd (p-waarde niet gerapporteerd)

Dronkers 2010

Onduidelijk

Postoperatieve complicaties, n:

I: 9/22

C: 8/20

Geen significant verschil tussen groepen (p=0.65)

Er waren geen adverse events in zowel de interventie- als controlegroep.

Kim 2009

Onduidelijk

Twee deelnemers in de interventiegroep vielen uit door vermoeidheid en algeheel onwelbevinden. Er waren geen andere adverse events (data niet weergegeven).

Moug 2019

De eerste paar weken na chirurgie

Postoperatieve complicaties, n:

I: 12/18

C: 12/22

Geen statistische toets uitgevoerd.

Deelnemers rapporteerden geen ernstige adverse events (data niet weergegeven).

Opnameduur

Carli 2010

Tot ontslag

Gemiddelde opnameduur, dagen (SD):

I: 11,9 (34,6)

C: 6,6 (3,6)

Onduidelijk of er een statistische toets werd uitgevoerd (p-waarde niet gerapporteerd)

Dronkers 2010

Tot ontslag

Opnameduur, dagen (SD):

I: 16,2 (11,5)

C: 21,6 (23,7)

Geen significant verschil tussen groepen (p=0.31)

Moug 2019

Tot ontslag

Mediane opnameduur, dagen (range):

I: 11 (6-37)

C: 10 (0-38)

Geen statistische toets uitgevoerd.

Kwaliteit van leven

Dronkers 2010

Tot chirurgie

Kwaliteit van leven vóór chirurgie (global health schaal van EORTC-QLQ-C30), score (SD):

I: 71 (19)

C: 68 (18)

Geen statistische toets uitgevoerd

Kwaliteit van leven vóór chirurgie (functionele schaal van EORTC-QLQ-C30), score (SD):

I: 413 (64)

C: 425 (67)

Geen statistische toets uitgevoerd

Kwaliteit van leven vóór chirurgie (symptoom schaal van EORTC-QLQ-C30), score (SD):

I: 119 (98)

C: 155 (117)

Geen statistische toets uitgevoerd

Verschilscore kwaliteit van leven (vóór chirurgie – baseline) (global health schaal van EORTC-QLQ-C30), score (SD):

I: 2 (19)

C: -3 (13)

Geen significant verschil tussen groepen (p=0.88)

Verschilscore kwaliteit van leven (vóór chirurgie – baseline) (functionele schaal van EORTC-QLQ-C30), score (SD):

I: 5 (56)

C: -2 (39)

Geen significant verschil tussen groepen (p=0.72)

Verschilscore kwaliteit van leven (vóór chirurgie – baseline) (symptoom schaal van EORTC-QLQ-C30), score (SD):

I: -35 (125)

C: 25 (75)

Geen significant verschil tussen groepen (p=0.20)

Moug 2019

1-2 weken vóór chirurgie

EORTC QLQ C30 (subschaal: vermoeidheid), gemiddelde score (SD) (verschilscore meting vóór chirurgie - baseline (95% betrouwbaarheidsinterval)):

I: 28.8 (23.9) (verschilscore: 0,7 (95%BHI: -13,7 tot 15))

C: 25.6 (16.9) (verschilscore: 7,2 (95%BHI: 1,8 tot 12,6))

Geen statistische toets uitgevoerd.

Overige subschalen van de EORTC QLQ C30 werden niet gerapporteerd.

Compliance

Dronkers 2010

Tot einde interventie

De aanwezigheid bij oefensessies voor de interventiegroep was 97%.

Kim 2009

Vóór chirurgie

De compliance in de interventiegroep was 74% (SD: 16%, range: 44-95%)

Patiënt-tevredenheid

Moug 2016

Einde van de trial

Tevredenheid over de trial (Likert schaalverdeling: 1 (helemaal niet mee eens) – 5 (helemaal mee eens)), mediane score (IQR):

I: 5 (5-5)

C: 4,5 (4-5)

Significant verschil tussen groepen (p=0.019).

Uitkomstmaat

Auteur

Duur follow-up

Resultaat

Morbiditeit, adverse events, en complicaties

Bousquet-Dion 2018

Tot 30 dagen na chirurgie

Tenminste één complicatie, n:

I: 14/37

C: 8/26

Geen significant verschil tussen groepen (p=0.562)

Gilis 2014

Tot 30 dagen na chirurgie

Ten minste één complicatie, n:

I: 12/38

C: 17/39

Geen significant verschil tussen groepen (p=0.277)

Opnameduur

Bousquet-Dion 2018

Tot 30 dagen na chirurgie

De lengte van de opnameduur was voor beide groepen vergelijkbaar (data niet weergegeven, p-waarde niet gerapporteerd).

Gilis 2014

Tot ontslag (zonder heropname)

Mediane opnameduur, dagen (IQR):

I: 4 (3-5)

C: 4 (3-7)

Geen significant verschil tussen groepen (p=0.812)

Kwaliteit van leven

Bousquet-Dion 2018

Tot 8 weken na chirurgie

Rapporteert in de methoden dat de kwaliteit van leven gemeten gaat worden, maar deze worden niet in de resultaten weergegeven.

Gilis 2014

Tot 8 weken na chirurgie

SF-36 (physiscal functioning): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.468)

SF-36 (role physical): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.360)

SF-36 (bodily pain): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.623)

SF-36 (general health): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.980)

SF-36 (vitality): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.597)

SF-36 (social functioning): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.656)

SF-36 (role emotional): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.672)

SF-36 (mental health): Geen verschil tussen groepen over de tijd tussen baseline en 8 weken postoperatief (p=0.085)

Compliance

Bousquet-Dion 2018

Tot aan chirurgie (preoperatief) en tot 8 weken postoperatief

Gemiddelde compliance van de preoperatieve lichamelijke oefeningen, %:

I: 98%

C: niet relevant

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddelde compliance van de preoperatieve voedingsadviezen, %:

I: 100%

C: niet relevant

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddelde compliance van de lichamelijke oefeningen 4 weken postoperatief, %:

I: 72%

C: 79%

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddelde compliance van de voedingsadviezen4 weken postoperatief, %:

I: 91%

C: 84%

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddelde compliance van de lichamelijke oefeningen 8 weken postoperatief, %:

I: 82%

C: 75%

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gemiddelde compliance van de voedingsadviezen 8 weken postoperatief, %:

I: 92%

C: 83%

Geen significant verschil tussen groepen (p-waarde niet gerapporteerd)

Gilis 2014

Tot 8 weken na chirurgie

Compliance (preoperatief), % (SD):

I: 78% (21%)

C: niet relevant

Compliance (tot 4 weken postoperatief), % (SD):

I: 53% (30%)

C: 31% (26%)

Significant verschil tussen groepen (p<0.001)

Compliance (van 4 tot 8 weken postoperatief), % (SD):

I: 53% (33%)

C: 40% (31%)

Geen significant verschil tussen groepen (p=0.117)

Study reference

(first author, publication year)

Describe method of randomisation¹

Bias due to inadequate concealment of allocation?²

(unlikely/likely/unclear)

Bias due to inadequate blinding of participants to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of care providers to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of outcome assessors to treatment allocation?³

(unlikely/likely/unclear)

Bias due to selective outcome reporting on basis of the results?⁴

(unlikely/likely/unclear)

Bias due to loss to follow-up?⁵

(unlikely/likely/unclear)

Bias due to violation of

intention to treat analysis?⁶

(unlikely/likely/unclear)

Bousquet-Dion 2018

Randomly assigned on a 1:1 ratio by computer generated random numbers

Unlikely

Reason: Random numbers were put in sealed envelopes (although not stated they were opaque)

Likely

Reason: participants cannot be blinded for the control or intervention procedures

Likely

Reason: health care providers cannot be blinded for the provided control or intervention procedures

Unlikely (primary outcome: walking capacity)

Reason: the assessor was blinded for group allocation.

Likely

Reason: From the methods it can be deduced that the quality of life was measured (SF-36, ED-5D), however it is not reported in the results. Results of the HADS (anxiety and depression) were only reported for the baseline and not for follow-up.

Likely

Reason: Loss to follow up varied between groups in proportion (I: 9.8%, C: 33,3%) and reasons.

Likely

Reason: Participants lost to follow-up were not included in the analyses. There is a difference in loss to follow-up proportion and reasons.

Likely (secondary outcomes)

Reason: it is likely that the healthcare providers were the outcome assessors of (most) secondary outcomes, such as complications and compliance.

Burden 2017

Participants were allocated on a 1:1 ratio by using blocks of 2. Allocation was stratified by tumour site and surgical approach. A statistician produced four lists of random numbers and an independent researcher set up the randomization procedure.

Likely

Reason: Although it was stated that the random numbers were kept in a sealed opaque envelopes, the block size of two is considered too small for allocation concealment. Only the first allocation of each block is concealed, where after the second allocation immediately is known.

Likely

Reason: authors state that the participants were not blinded for the interventions.

Unlikely

Reason: the supplement and water were packaged in the same type of sealed carton box with identical weight. Boxes were given to participants. Authors state that the research team was blinded for the interventions.

Unlikely

Reason: the supplement and water were packaged in the same type of sealed carton box with identical weight. Boxes were given to participants. Authors state that the research team was blinded for the interventions.

Unlikely

Reason: no indication of selective reporting.

Unlikely

Reason: only 1 participant was lost during follow-up

Unlikely

Reason: ITT was followed for n=2 in each group who did not receive the allocated intervention. The participant lost in the follow-up was excluded from analyses. Chance of bias due to 1 person lost in follow-up is minimal.

MacFie 2000

Patients were randomized in groups (random sequence generation procedures not described)

Unclear

Reason: procedures not described

Likely

Reason: Authors state that the trial was not blinded.

Likely

Reason: Authors state that the trial was not blinded.

Likely

Reason: Authors state that the trial was not blinded.

Likely

Reason: Authors state in their methods that a quality of life questionnaire was administered, however no quality of life results were reported.

Unlikely

Reason: n=12 were removed from the study because surgery was cancelled or urgent surgery was necessary. No other drop-out during the course of the study.

Unlikely

Reason: no indication for protocol deviations.

Moug 2019

Randomization in 1:1 ratio by using block size 4, without stratification. An interactive voice response telephone system was used to inform the research assistant about the

Unlikely

Reason: A research assistant was informed about the allocation by a computerized telephone system, and provided the allocation to the participant

Likely

Reason: participants cannot be blinded for the intervention they received.

Unlikely:

Authors state that the testers, surgeons, nurse specialists, and hospital staff were blinded for group allocation

Unlikely:

Authors state that the testers, surgeons, nurse specialists, and hospital staff were blinded for group allocation

Likely

Reason: Only two scales of the EORTC QLQ C30 (quality of life) were reported at baseline, and only one scale was reported in the follow-up.

Likely

Reason: Loss to follow up varied between groups in proportion (I: 25%, C: 8,3%) and reasons.

Likely

Reason: ITT was not followed. There is a difference in proportions of and reasons for drop-out. Furthermore there are participants excluded from some analyses (probably due to missing data).

Sorensen 2014

Randomization was performed (number/sequence generation procedures not described)

Unlikely

Reason: allocation was concealed by sealed opaque envelopes. The sealed envelopes were stored at the investigation site.

Unlikely

Reason: the provided cartons for the intervention and control group had identical looks, taste, and scent. Authors state that patients were blinded for treatment.

Unlikely

Reason: the provided cartons for the intervention and control group had identical looks, taste, and scent. Authors state that care providers were blinded for treatment.

Unlikely

Reason: the provided cartons for the intervention and control group had identical looks, taste, and scent. Authors state that researchers were blinded for treatment.

Unlikely

Reason: No indication for selective outcome reporting based on the manuscript and the outcomes of interest.

Unlikely

Reason: loss to follow up is moderate in size and similar in numbers and reasons for both groups.

Unlikely

Reason: ITT was followed.