Low intra-abdominal pressure in laparoscopic surgery: a systematic review and meta-analysis

Background:

Guidelines do not provide clear recommendations with regard to the use of low intra-abdominal pressure (IAP) during laparoscopic surgery. The aim of this meta-analysis is to assess the influence of low versus standard IAP during laparoscopic surgery on the key-outcomes in perioperative medicine as defined by the StEP-COMPAC consensus group.

Materials and methods:

We searched the Cochrane Library, PubMed, and EMBASE for randomized controlled trials comparing low IAP (<10 mmHg) with standard IAP (10 mmHg or higher) during laparoscopic surgery without time, language, or blinding restrictions. According to the PRISMA guidelines, two review authors independently identified trials and extracted data. Risk ratio (RR), and mean difference (MD), with 95% CIs were calculated using random-effects models with RevMan5. Main outcomes were based on StEP-COMPAC recommendations, and included postoperative complications, postoperative pain, postoperative nausea and vomiting (PONV) scores, and length of hospital stay.

Results:

Eighty-five studies in a wide range of laparoscopic procedures (7349 patients) were included in this meta-analysis. The available evidence indicates that the use of low IAP (<10 mmHg) leads to a lower incidence of mild (Clavien–Dindo grade 1–2) postoperative complications (RR=0.68, 95% CI: 0.53–0.86), lower pain scores (MD=−0.68, 95% CI: −0.82 to 0.54) and PONV incidence (RR=0.67, 95% CI: 0.51–0.88), and a reduced length of hospital stay (MD=−0.29, 95% CI: −0.46 to 0.11). Low IAP did not increase the risk of intraoperative complications (RR=1.15, 95% CI: 0.77–1.73).

Conclusions:

Given the established safety and the reduced incidence of mild postoperative complications, lower pain scores, reduced incidence of PONV, and shorter length of stay, the available evidence supports a moderate to strong recommendation (1a level of evidence) in favor of low IAP during laparoscopic surgery.

Introduction

Highlights

• Low pressure laparoscopy reduces early pain scores, postoperative nausea and vomiting (PONV) and length of stay.

• Low pressure laparoscopy reduces the risk of mild postoperative complications.

• Low pressure laparoscopy does not increase the rate of intraoperative complications.

• The use of low intra-abdominal pressure (IAP) during laparoscopic surgery is recommended.

The Enhanced Recovery After Surgery (ERAS) guidelines constitute a major advancement in evidence-based improvement of postoperative patient outcomes. Continuous updating of these guidelines with the latest clinical trial evidence allows for the most advanced level of patient care. One of the elements included in many of the ERAS Society protocols is minimally invasive or laparoscopic surgery, advocated for fewer complications and a quicker recovery. While the European Association for Endoscopic Surgery consensus guidelines advises using the lowest IAP with an adequate view of the surgical field1, the ERAS Society seems to recognize the importance of IAP but is conservative in their current recommendations. The ERAS guidelines for gastrointestinal2,3, colorectal4, gynecologic5 and bariatric6 surgery all discuss some strategies as neuromuscular blockade allows for lower insufflation pressures, but no target level of pressure is explicitly recommended. For more specialized procedures such as gastrectomy7, liver resection8, esophagectomy9 and pancreatoduodenectomy10, evidence regarding minimally invasive surgery is limited and IAP is not mentioned. Heterogeneity between guidelines due to inconsistent evidence concerning the safety and efficacy of low pneumoperitoneum pressure probably explains the large variability in routine clinical practice. The standard practice in most centers, is set the IAP to a routine pressure of 12–15 mmHg11 instead of the lowest possible IAP with an adequate view of the surgical field. Although a consensus definition is lacking, we predefined low pneumoperitoneum pressure as an IAP less than 10 mmHg in line with the described perfusion pressure in the distal segment of the capillary network of the parietal peritoneum12. Moreover, a sustained increase in IAP of 12 mmHg or higher is defined as intra-abdominal hypertension and could lead to organ ischemia13–15. There is physiological evidence that low IAP reduces postoperative pain scores16,17 and recent studies from our group show that acute pain in the first hours to days after surgery correlates with 30-day complications18,19. While a causal relationship between early pain scores and complications after surgery has not been fully established, it is compelling to hypothesize that strategies that decrease early postoperative pain may reduce complications after surgery. In a previous systematic review and meta-analysis16, postoperative complications could not be included as outcome measure due to insufficient reporting. Also, sufficient data with regard to safety and other clinical outcomes were lacking. Moreover, only relatively small studies with high/unclear risk of bias were available for the meta-analysis published in 2016. Therefore, an updated analysis of the available clinical evidence concerning the impact of the IAP during laparoscopic surgery is warranted.

The aim of this systematic review and meta-analysis was to assess the influence of low IAP during laparoscopic surgery on clinical and patient core outcomes in perioperative care as defined by the Standardized Endpoints in Perioperative medicine – Core Outcome Measures for Perioperative and Anesthetic Care (StEP-COMPAC) consensus group20.

Materials and methods

Search strategy

This review was conducted according to the PRISMA guidelines recommended by the Cochrane Handbook21, see PRISMA checklist in Supplement (Supplemental Digital Content 2, http://links.lww.com/JS9/A270). The PubMed, EMBASE, and Cochrane databases were searched without time, language and blinding restrictions. Thereafter, references and cross-references were searched by hand. The last search was carried out in November 2021. The databases were searched with ‘laparoscopy,’ ‘peritoneoscopy,’ ‘coelioscopy,’ and ‘celioscopy,’ and with ‘pneumoperitoneum,’ ‘artificial pneumoperitoneum,’ and ‘insufflation’ in Mesh term with an exploration of all trees and of titles or abstracts with word variations. This resulted in the following search strategies:

In PubMed: (laparoscop* OR coelioscop* OR celioscop* OR peritoneoscop*) AND (pneumoperitoneum OR pneumoperitoneum, Artificial [MeSH] OR insufflations OR insufflation [MeSH] AND (randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized [tiab] OR randomly [tiab] OR trial [tiab]).

In EMBASE: (exp Laparoscopy/ OR laparoscop*.ti,ab,kw. OR peritoneoscop*.ti,ab,kw. OR coelioscop*.ti,ab,kw. OR celioscop*.ti,ab,kw.) AND (Pneumoperiton*.ti,ab,kw. OR Pneumo-periton*.ti,ab,kw. OR artificial pneumoperitoneum/ OR Pneumoperitoneum/ OR insufflation/).

In Cochrane library: (MeSH [Laparoscopy] OR (laparoscop* OR peritoneoscop* OR coelioscop* OR celioscop*):ti,ab,kw OR MeSH [pneumoperitoneum] OR [pneumoperitoneum, artificial] OR [insufflation] OR (Pneumoperiton* OR Pneumo-periton*):ti,ab,kw.

Data outcomes

The StEP-COMPAC consensus defined some clinical and patient-centered core outcomes divided over six domains for use in perioperative clinical trials20. Mortality/survival; perioperative complications; resource use; short-term recovery after surgery; longer term recovery after surgery and overall success/failure of surgery. In line with StEP-COMPAC, the main outcomes of this meta-analysis included the core outcomes of all domains.

In addition, we included secondary outcomes related to patient safety and surgical procedural feasibility and success, including intraoperative complications, quality of the surgical working field (quantified with the Leiden Surgical Rating Scale), the incidence of conversion to laparotomy or higher IAP, duration of surgery and blood loss. Other collected data were mean age, sex, BMI, type of procedure, depth of neuromuscular block (NMB), incidence of shoulder pain, analgesia use (morphine equivalent), heart rate, mean arterial blood pressure, end-tidal CO₂, liver enzymes, and indicators of inflammation of immune suppression. For missing or unclear information, authors of the original articles were contacted. If no reply was received, SDs were calculated if sufficient data were available22.

Eligibility criteria

All randomized controlled trials in human adults comparing IAPs lower than 10 versus 10 mmHg or higher, for standard or robotic laparoscopic intra-abdominal, intraperitoneal procedures were eligible. Studies published in all languages were included and translated if applicable.

Selection process and data extraction

All search results were imported in Covidence, duplicates were removed. Dual screening and data collection were performed independently by two reviewers per article (G.R. and L.J.). Conflicts were resolved by a third reviewer (K.A. or M.W.) for the final inclusion decision. Thereafter, all data was cross-checked. Continuous variables were imported as mean with standard deviation, categorical variables were imported as number of events and total. For missing or inconclusive data, the authors were contacted for clarification or additional data.

Risk of bias and certainty of evidence assessment

For the quality assessment of randomized controlled trials, the Cochrane Risk of Bias tool was used21. Two independent reviewers (G.R. and L.J.) judged the risk of bias using this tool. Conflicts were resolved with a third independent reviewer (M.W.). No automated tools were used.

Grading of the quality of evidence was performed using the Cochrane adapted GRADE approach by two reviewers (G.R. and M.W.), which places outcomes in one of four levels of certainty of evidence; high, moderate, low and very low23.

Calculation and statistical methods

Studies comparing an IAP of lower than 10 versus 10 mmHg or higher were included (Table 1). Data from studies with more than one group in one of these study arms were combined in Review Manager (version 5.3, The Cochrane Collaboration) as instructed by the Cochrane handbook21. For example, an included study with three groups randomized in IAP of 8, 12, or 16 mmHg. The data of the last two groups were combined for use in the 10 mmHg and higher study arm of this meta-analysis. Missing data were calculated according to Cochrane guidelines where possible.

Table 1.

Characteristics of included studies

	IAP (mmHg)				Number of patients
References	Low	Standard	Deep NMB	Total number of patients	Low	Standard	Type of procedure	Poor qualitya
Aditianingsih et al. 24	8	12	No	44	22	22	LDN
Akkoc et al. 25		10/12/14	Unknown	76		24/25/27	UL	Yes
Albers et al. 26	8	12	Yesb	178	89	89	LC
Ali et al. 27		10/>10	Unknown	160		80/80	LChol	Yes
Anwaar et al. 28	7–10	12–16	Unknown	100	50	50	LChol	Yes
Aydin et al. 29	8	12	Unknown	60	30	30	LChol
Barczynski et al. 30	7	12	Unknown	148	74	74	LChol
Barrio et al. 31	8	12	Yesc	90	60	30	LChol
Basgul et al. 32		10/14–15	Unknown	22	11	11	LChol	Yes
Bhattacharjee et al. 33	9–10	14	Unknown	80	40	40	LChol
Bogani et al. 34	8	12	Unknown	42	20	22	LH
Budhiraja et al. 35	7	12	Unknown	100	50	50	LChol	Yes
Cai et al. 36		10/12/15	Unknown	66		22/22/22	LC
Celarier et al. 37	7	12	Yesd	127	62	65	LC
Celik et al. 38	8	12/14	Unknown	60	20	20/20	LChol
Chang et al. 39	6–8/9–11	12–14	Unknown	150	50/50	50	LChol
Chok et al. 40	7	12	Unknown	40	20	20	LChol
de’Angelis et al. 41	8	12	Unknown	161	35	126	LChol	Yes
Dexter et al. 42	7	15	Unknown	20	10	10	LChol	Yes
Diaz-Cambronero et al. 43	8	12	Yesb	166	85	81	LC
Doğan et al. 44	7	10/13	Unknown	90	30	30/30	LChol
Donmez et al. 45		10/14	Unknown	50		25/25	LChol
Ekici et al. 46	7	15	Unknown	52	20	32	LChol
Eryilmaz et al. 47		10/14	Unknown	43		20/23	LChol
Esmat et al. 48		10/14	Unknown	71		34/37	LChol	Yes
Gin et al. 49	8	12	No	100	51	49	LChol
Goel et al. 50	7–10	12–14	Unknown	60	30	30	LChol	Yes
Gupta et al. 51	8	14	Unknown	101	50	51	LChol
Hasukić et al. 52	7	14	Unknown	50	25	25	LChol
Ibraheim et al. 53	6–8	12–14	No	20	10	10	LChol	Yes
Joshipura et al. 54	8	12	Unknown	26	14	12	LChol
Kandil et al. 55	8	10/12/14	Unknown	100	25	25/25/25	LChol	Yes
Kanwer et al. 56	7–10	12–14	Unknown	60	30	30	LChol	Yes
Karagulle et al. 57	8	12/15	No	45	15	15/15	LChol	Yes
Kendir et al. 58		10/14	Unknown	40		20/20	LChol	Yes
Khan et al. 59	8	14	Unknown	214	107	107	LChol
Kim et al. 60	8	13	No	46	23	23	GL
Kim et al. 61	8	12	Yesb	74	37	37	GL
Koc et al. 62		10/15	Unknown	50		25/25	LChol
Ko-Iam et al. 63	7	14	Unknown	120	60	60	LChol
Kundu et al. 64	8	10/12/15	Unknown	360	62	102/123/73	GL
Luo et al. 65		10/15	Unknown	102		51/51	RL	Yes
Madsen et al. 66	8	12	Yese	28	14	14	GL
Madsen et al. 67	8	12	Yesb	99	49	50	LH
Madsen et al. 68	8	12	Yesb	110	55	55	LH
Marton Filho et al. 69	6–8	10–12	Yesd	64	31	33	LChol
Matsuzaki et al. 70	8	12	Unknown	68	32	36	LH	Yes
Matsuzaki et al. 71	8	12	No	82	41	41	LH
Mohammadzade et al. 72	7–10	12–14	Unknown	60	30	30	LChol	Yes
Moro et al. 73		10/14	Yesd	80		40/40	LChol
Murtaza et al. 74	8	12	Unknown	80	40	40	LChol	Yes
Nasajiyan et al. 75	7–9	14–15	Unknown	50	25	25	LChol
Nematihonar et al. 76	6–8	12–14	Unknown	202	101	101	LChol	Yes
Neogi et al. 77	7	14	Unknown	80	32	48	LChol
Nuna et al. 78	8–10	12–14	Unknown	50	25	25	LChol	Yes
O et al. 79	7/9	12	No	54	20/21	13	LChol	Yes
Özdemir-van Brunschot et al. 80	6	12	Yesd	63	33	30	LDN
Perrakis et al. 81	8	15	Unknown	40	20	20	LChol
Pulle et al. 82	8–10	13–15	Unknown	194	94	100	LChol
Radosa et al. 83	8	15	Unknown	178	91	87	LH
Rehman et al. 84	7	14	Unknown	60	30	30	LChol	Yes
Rohloff et al. 85	8	12	Unknown	201	96	105	RALP
Rosenberg et al. 86	8	12	Yesf	127	60	67	LChol
Sandhu et al. 87	7	14	Unknown	140	70	70	LChol	Yes
Sandoval-Jiménez et al. 88	7	12–15	Unknown	68	34	34	LChol
Sarli et al. 89	9	13	Unknown	90	46	44	LChol
Sattar et al. 90		<12/12–16	Unknown	180	90	90	LChol	Yes
Schietroma et al. 91	6–8	12–14	Unknown	68	33	35	LNF	Yes
Schietroma et al. 92	6–8	12–14	Unknown	51	25	26	LA
Sefr et al. 93		10/14	Unknown	32		10/22	LChol
Sharma et al. 94	8	14	Unknown	50	25	25	LChol
Shoar et al. 95	8	12	Unknown	50	25	25	LChol
Singla et al. 96	7–8	12–14	No	100	50	50	LChol
Song et al. 97		10/11–12/13–14/15–16	Unknown	118		35/31 /28/24	GL	Yes
Sood et al. 98	8–10	15	Yesd	9	5	4	LA
Sroussi et al. 99	7	15	Unknown	60	30	30	GL
Topal et al. 100		10/13/16	Unknown	60		20/20/20	LChol
Topçu et al. 101	8	12/15	Unknown	150	54	48/48	GL
Torres et al. 102	6–8	12–14	Unknown	40	20	20	LChol	Yes
Vijayaraghavan et al. 103	8	12	No	43	22	21	LChol
Wallace et al. 104	7.5	15	Unknown	40	20	20	LChol
Warlé et al. 105	7–9	12–16	No	20	10	10	LDN
Xiao et al. 106	7–9	11–13	Unknown	96	48	48	LChol	Yes
Yasir et al. 107	8	14	Unknown	100	50	50	LChol	Yes
Zaman et al. 108	7–8	12–14	Unknown	50	25	25	LChol	Yes

GL, gynecological laparoscopy; IAP, intra-abdominal pressure; LA, laparoscopic adrenalectomy; LC, laparoscopic colorectal surgery; LChol, laparoscopic cholecystectomy; LDN, laparoscopic donor nephrectomy; LH, (mini-) laparoscopic hysterectomy; LNF, laparoscopic Nissen fundoplication; NMB, neuromuscular block; RALP, robot-assisted laparoscopic prostatectomy; RL, retroperitoneal laparoscopic surgery; UL, urinary tract laparoscopy.

^a Assessed according to the Cochrane risk of bias Tool; three or more bias items scored ‘unclear bias’ or ‘high risk’.

^b Deep NMB with low IAP and moderate NMB with standard IAP.

^c Deep and moderate NMB in low group.

^d Deep NMB in both groups.

^e Deep NMB and no NMB in both groups.

^f Deep and moderate in both groups.

Meta-analysis was performed for all outcomes where sufficient data were available. We reported risk ratio (RR) and mean difference (MD) for dichotomous and continuous outcomes, respectively. The presence and extent of heterogeneity was measured by χ² and I ² calculated by Review Manager. Given the expected heterogeneity between the studies, a random-effects model was used for all analyses.

Sensitivity analyses were performed for (1) exclusion of all poor quality studies (a study was classified as ‘poor quality’ when three or more risk of bias items scored ‘unclear bias’ or ‘high risk bias’), (2) exclusion of laparoscopic cholecystectomy to assess the effects for longer and/or more complex procedures, (3) only studies where deep NMB was used to facilitate the use of low IAP, and (4) with an IAP of 10 mmHg categorized as low pressure instead of standard pressure, as the precise definition of low pneumoperitoneum pressure is lacking.

Results

Study selection and characteristics

Database’s search is depicted in the PRISMA flowchart (Fig. 1). A total of 8584 references were found and imported in Covidence and 3200 duplicates were removed. After review and screening 489 studies where selected for full text review. Finally, 85 studies were included in the analysis. Study characteristics of the included studies24–108 are presented in Table 1. There are thirty poor quality studies according to the risk of bias assessment25–28,32,35,41,42,48,50,53,55–58,65,70,72,74,76,78,79,84,87,90,91,97,102,106–108.

Figure 1.

PRISMA flowchart.

Main outcomes

Despite careful evaluation there were only available data for perioperative complications, resource use and short-term recovery after surgery. The main outcomes were postoperative complications (in hospital and within 30 days after surgery), postoperative pain (up to 72 h postoperative), PONV, and length of hospital stay. There was no available data for three domains: mortality/survival, longer term recovery after surgery, and overall success/failure of surgery.

Table 2A presents the summary of findings for the StEP-COMPAC core outcomes, starting with the main outcomes of this review. At low IAP (<10 mmHg), patients developed significantly less minor postoperative complications (RR=0.60, 95% CI: 0.46–0.78, P=0.001). This significant difference was not present for grade 3 and 4 complications, which are considered major complications according to the StEP-COMPAC criteria. Postoperative pain was significantly lower at low IAP (MD=−0.68, 95% CI: −0.82 to 0.54, P<0.00001). PONV were significantly lower for low IAP over all time points combined (RR=0.67, 95% CI: 0.51–0.88, P 0.004). Length of hospital stay (in days) was significantly shorter with low IAP (MD=−0.29, 95% CI: −0.46 to −0.11; P<0.001). The forest plots of postoperative complications, length of hospital stay, postoperative pain, and PONV are presented in Figure 2. For the remaining StEP-COMPAC core outcomes, insufficient data were available for meta-analysis.

Table 2.

Summary of findings table

Patients	Individuals undergoing laparoscopic surgery
Intervention	Low intra-abdominal pressure (<10 mmHg)
Comparison	Standard intra-abdominal pressure (≥10 mmHg)
Outcome		Number of participants (studies)	Relative effects (95% CI)	P	I 2 (%)	Certainty of the evidence (GRADE)	Comments
2A: Main outcomes20
Mortality/survival
Overall mortality		No data	NA	NA	NA	NA
Long-term survival		No data	NA	NA	NA	NA
Perioperative complications
(Major) Postoperative complications
CD grade 1–2		1597 (13)	RR=0.60 (0.46–0.78)	0.0001	0	⊕⊕⊕⊕ High
CD grade 3–4		1507 (12)	RR=1.25 (0.71–2.20)	0.44	0
No classification		290 (6)	RR=0.40 (0.05–3.34)	0.40	10
Total		3394 (21)	RR=0.68 (0.53–0.86)	0.001	0
Complications with permanent disability		No data	NA	NA	NA	NA
Resource use
Hospital stay		2562 (22)	MD=−0.29 (−0.46 to −0.11)	0.001	89	⊕⊕⊕⊝ Moderate Due to inconsistency	Substantial heterogeneity
Hospital readmission		No data	NA	NA	NA	NA
Short-term recovery after surgery
Postoperative Pain (NRS)
0–12 h		2355 (24)	MD=−0.72 (−0.96 to −0.47)	<0.00001	55	⊕⊕⊕⊝ Moderate Due to inconsistency	Substantial heterogeneity
12–24 h		1977 (23)	MD=−0.72 (−0.98 to −0.47)	<0.00001	73
24–72 h		1389 (16)	MD=−0.60 (−0.86 to −0.35)	<0.00001	66
Total		5721 (27)	MD=−0.68 (−0.82 to −0.54)	<0.00001	67
PONV (yes/no)
0–12 h		434 (5)	RR=0.80 (0.55–1.17)	0.25	44	⊕⊕⊕⊝ Moderate Due to inconsistency	Moderate heterogeneity
12–24 h		238 (2)	RR=0.61 (0.18–2.13)	0.44	69
24–72 h		238 (2)	RR=0.46 (0.19–1.14)	0.09	0
No timestamp		429 (5)	RR=0.59 (0.39–0.90)	0.01	29
Total		1339 (10)	RR=0.67 (0.51–0.88)	0.004	42
Psychological wellbeing		No data	NA	NA	NA	NA
Discharge destination		No data	NA	NA	NA	NA
Longer term recovery after surgery
Overall health-related quality of life		No data	NA	NA	NA	NA
Overall success/failure of surgery
Patient satisfaction		No data	NA	NA	NA	NA
2B: Safety outcomes surgical procedural feasibility and success
Intraoperative complications		1661 (16)	RR=1.15 (0.77–1.73)	0.50	0	⊕⊕⊕⊕ High
Quality of the surgical fielda		896 (10)	MD=−0.63 (−1.14 to −0.13)	0.01	97	⊕⊕⊕⊝ Moderate Due to inconsistency	Considerable heterogeneity
Conversion
To laparotomy		1718 (20)	RR=1.36 (0.55–3.33)	0.50	0	⊕⊕⊕⊕ High
To higher pressure		1411 (16)	RR=4.71 (2.88–7.69)	<0.00001	0
Duration of surgery (min)		5047 (55)	MD=1.75 (0.15–3.64)	0.07	89	⊕⊕⊕⊝ Moderate Due to inconsistency	Substantial heterogeneity
Blood loss (ml)		861 (8)	MD=16.30 (−9.40 to 41.99)	0.21	92	⊕⊕⊕⊝ Moderate Due to inconsistency	Considerable heterogeneity

CD, Clavien–Dindo; MD, mean difference; NA, not available; NRS, Numerical Rating Scale; PONV, postoperative nausea and vomiting; RR, risk ratio.

^a Scored on the Leiden Surgical Rating Scale: 5=excellent, 4=good, 3=improvement is welcome, 2=bad, 1=very.

Figure 2.

Forest plots of main outcomes20. (A) Postoperative complications by Clavien–Dindo grade. (B) Postoperative pain. (C) Postoperative nausea and vomiting (PONV). (D) Length of hospital stay.

For the secondary safety outcomes (reported in Table 2B), there was a statistically significant difference for low IAP regarding quality of the surgical field (better for standard pressure). Intraoperative complications, conversion to laparotomy, operation time and blood loss showed no statistically significant differences. The summary of findings table of all other outcomes are presented in the Supplementary (Supplementary Table 3, Supplemental Digital Content 2, http://links.lww.com/JS9/A270), including the forest plots (Supplementary Fig. 1, Supplemental Digital Content 2, http://links.lww.com/JS9/A270).

The funnel plots of all outcomes are presented in the Supplementary Figure 2 (Supplemental Digital Content 2, http://links.lww.com/JS9/A270) and do not indicate publication bias for the main and secondary outcomes. Heterogeneity between studies was present as displayed in Table 2. Postoperative complications had a low I ², PONV showed a moderate heterogeneity on average by mostly staying below the margin of 60%. Postoperative pain scores and hospital stay had substantial heterogeneity with I ² above 60%.

Sensitivity analyses did not indicate a significant influence of poor quality studies or studies investigating laparoscopic cholecystectomy (Supplementary Table 2, Supplemental Digital Content 2, http://links.lww.com/JS9/A270). Differences in favor of low IAP defined as less than 10 mmHg or lower were no longer significant for all four main outcomes when 10 mmHg was categorized as low pressure. The number of studies explicitly reporting the use of deep NMB to facilitate low pressure was too small to perform meaningful sensitivity analyses.

Risk of bias and certainty of evidence

The risk of bias assessment presented the highest risk of bias in incomplete outcome data, although with the high risk and unclear categories taken together, this occurred in less than 50% of all studies (Fig. 3). Publication bias was limited based on the funnel plots (Supplementary Fig. 2, Supplemental Digital Content 2, http://links.lww.com/JS9/A270). The risk of bias per study is presented in detail in Table 1 of the Supplementary (Supplemental Digital Content 2, http://links.lww.com/JS9/A270).

Figure 3.

Risk of bias summary: judgement of each risk of bias criterion across all included studies.

Table 2 shows the certainty of evidence per outcome, including reasons for downgrading. This was mostly based on heterogeneity between studies, which is reflected by the diversity of surgery types included in this meta-analysis. Because of a limited number of studies, almost all biochemical outcomes were considered imprecise.

Discussion

The main results of this meta-analysis show that the use of low IAP significantly reduces the incidence of mild to moderate postoperative complications (Clavien–Dindo grade 1–2); reduces 24–72 h postoperative pain scores; reduces PONV scores; and reduces the mean length of hospital stay. The overall certainty of evidence according to the GRADE approach was moderate to high. Sensitivity analyses indicated that the mean differences were similar in patients undergoing laparoscopic cholecystectomy as compared with other laparoscopic procedures. This indicates that the benefits of lower IAP during laparoscopic surgery are not procedure-bound.

Over the last years, a relatively large number of randomized clinical trials had studied the influence of low IAP during laparoscopic surgeries. Our quality assessment revealed that a vast majority of these trials (55/85) displayed a low risk of bias. Our findings are in line with a previous meta-analysis from our group including 33 randomized controlled trials mainly studying patients undergoing laparoscopic cholecystectomy16. In this previous study we demonstrated that the use of low IAP reduced pain scores 24–48 h after surgery as compared with standard IAP. Up until the meta-analysis of Gurusamy and colleagues on low versus standard IAP in laparoscopic cholecystectomy patients in 2014, there was a lack of evidence and no meaningful conclusions could be drawn regarding efficacy outcomes due to high risk of bias of the studies available. Because the number of studies published on low versus standard pressure laparoscopy with low risk of bias increased steadily, a reliable meta-analysis could be performed for postoperative pain intensity and shoulder pain incidence as efficacy outcome measures in 201516. The number of studies further increased after 2015, and the reporting of other clinically relevant outcomes improved which allowed for an update of our previous meta-analysis. There were no studies reporting on mortality, and longer term recovery after surgery. Nevertheless, sufficient data were available with regard to four important outcome measures in perioperative medicine, as defined by the StEP-COMPAC consensus group (i.e. postoperative complications, postoperative pain, PONV and length of hospital stay). In the hierarchy of outcome measures, mortality, and severe postoperative complications are considered more important for guideline decision-makers than minor postoperative complications, postoperative pain, and PONV. Nevertheless, an intervention which improves short-term recovery only, may still be important to adopt when considered safe.

In summary, the current meta-analysis provides high certainty of evidence indicating that the use of low pressure does not increase the incidence of intraoperative complications, while reducing the incidence of Clavien–Dindo grade 1–2 postoperative complications among other benefits in clinical and patient-centered outcomes. In our view, this strongly supports the safety of using low IAP. The use of deep NMB could in theory facilitate the use of lower IAP during laparoscopic surgery, by increasing workspace and reducing muscle contractions. A meta-analysis by Bruintjes et al. 109 confirmed that deep NMB improved surgical conditions rated by surgeons during laparoscopic surgery as compared with moderate NMB. The number of studies reporting on deep NMB to facilitate low pneumoperitoneum pressure was too small to allow for a meaningful subgroup analysis regarding relevant efficacy endpoints. Therefore, further studies are warranted to establish the optimal depth of muscle relaxation for low pressure during laparoscopy. Next to adequate muscle relaxation, prestretching of the abdominal wall musculature with a short period of standard IAP (12–15 mmHg) during trocar introduction, could also contribute to increased working space during subsequent low IAP26,43. Although, poorly described in a vast majority of studies included in this meta-analysis, several studies used standard insufflation pressures during trocar insertion to create some prestretching of the abdominal wall. It seems unlikely that a short period of prestretching with standard IAP compromises the clinical benefits of low IAP, but further studies are required to confirm this.

This meta-analysis provides high certainty evidence for the safety of low IAP during laparoscopy as reflected by a similar incidence of intraoperative complications. The use of low pressure (regardless of NMB) did modestly reduce the quality of the surgical field with 0.6 points on a five-point scale. But the clinical benefits of low IAP are reflected by a significant, but modest reduction in postoperative pain, PONV, and length of hospital stay. These modest benefits combined with a reduced incidence of postoperative complications (mild to moderate, Clavien–Dindo grade 1–2 complications) justify a moderate to strong recommendation in favor of using low IAP (<10 mmHg) during laparoscopic surgery. It is important to stress that surgeons should increase the IAP when surgical conditions are inadequate according to their judgement. Although, a higher IAP does not always improve the surgical conditions probably due to patient bound factors. Deep NMB could enhance surgical conditions and should be considered when using low IAP109.

Sensitivity analyses in which patients allocated to an insufflation pressure of 10 mmHg were added to the low pressure group, failed to demonstrate a significant beneficial impact of low IAP regarding the main outcomes. In our view, this finding supports the main hypothesis in a vast majority of the existing studies in which clinical benefits were expected by using a pressure of less than 10 mmHg or lower, instead of 10 mmHg or lower. Therefore, laparoscopic surgeons aiming for better recovery of their patients with a lower risk of mild complications should use an insufflation pressure below 10 mmHg, but – to ensure safety – low IAP should only be used when the quality of the surgical field is adequate according to their judgement.

The main strength of this meta-analysis is its solid approach and methods in line with the handbook of the Cochrane collaboration. The high number of studies available allowed us to perform meaningful analyses for a wide range of outcome measures. Although moderate to substantial heterogeneity was present for most of the outcomes, the certainty of evidence for these outcomes was moderate to high. Ideally, subgroup analyses would have been performed for each type of laparoscopic procedure, since the ease to work with low IAP could differ between surgery taking place high abdominal or in small pelvis. But – except for laparoscopic cholecystectomy – the available number of studies was too small. Nevertheless, according to Pascal’s law carbon dioxide pressure in the abdominal cavity equally expands in all directions, and therefore it is reasonable to assume that the physiological impact of a pneumoperitoneum does not differ significantly between different types of procedures. Despite this, low IAP may be more difficult to use in certain types of procedures depending on the complexity of the procedure, positioning, and patient-related factors (i.e. obesity, previous surgery). Also, the use of low IAP might, at least in theory, be easier in robot-assisted procedures with greater ranges of motion and more precision in narrow working spaces.

In our view it may not be safe to use low IAP for blind trocar introduction, through the umbilical incision, after insufflation with a Verres needle. Also, during trocar introduction under camera vision it may be better to use standard insufflation pressures, to minimize the risk of trocar injuries, but also to induce a certain degree of prestretching of the abdominal wall muscles. Some reports demonstrate that prestretching facilitates the use of low IAP after trocar introduction110,111. Therefore, our recommendation to use low IAP does not refer to the initial trocar introduction phase. To guarantee optimal safety, and to create some prestretching of the abdominal wall, an insufflation pressure of 12–15 mmHg during trocar introduction would, at least in theory, be ideal. Once the initial trocars have been introduced, the IAP can be reduced below 10 mmHg. However, if surgical conditions remain poor according to the surgeon’s judgement after optimization of positioning and neuromuscular blockade, it is not safe to continue with low IAP.

Limitations of this study are mainly related to the relatively high number of poor quality studies, and the high incidence of poorly reported outcome measures. Nevertheless, due to the high number of studies available we could perform a sensitivity analysis in which poor quality studies were excluded. The results regarding the main outcomes were not influenced by the presence of poor quality studies. Finally, a relatively high number of comparisons were carried out in this meta-analysis without correction for multiplicity. However, this approach is in line with the Cochrane handbook which explicitly does not recommend adjustments for multiple comparisons.

In conclusion, regarding the core outcomes in perioperative medicine, the use of low IAP leads to a lower incidence of mild postoperative complications, lower pain scores and PONV, and a slightly reduced length of hospital stay. Also, the use of low IAP in a wide range of laparoscopic procedures is associated with a slightly reduced quality of the surgical field, however high certainty evidence does not indicate an increased risk of intraoperative complications or conversion to laparotomy.

Therefore, our data support a moderate to strong recommendation (1a level of evidence) in favor of the use of low IAP during laparoscopic surgery. It is important to note that surgeons should be aware that IAP should be increased when the quality of the surgical field is inadequate to ensure safety.

Ethical approval

NA.

Sources of funding

There was no financial or nonfinancial support used or related to the process or writing of this manuscript.

Author contribution

G.T.J.A.R.-B., E.v.H., and M.C.W. were involved in the study design. G.T.J.A.R.-B., K.I.A., E.v.H., and L.M.C.J. collected data. Interpretation of the data was done by G.T.J.A.R.-B., L.M.C.J., and M.C.W. with substantial input of O.D.-C., G.M., C.R., G.-J.S., and C.K. In the writing process, G.T.J.A.R.-B., M.C.W. were essential. Substantial input in the writing process and revisions were done by all authors.

Conflicts of interest disclosure

M.C.W. and O.D.-C. received grants from Merck Sharp & Dohme for investigator-initiated studies. The remaining authors declare that they have no financial conflict of interest with regard to the content of this report.

Research registration unique identifying number (UIN)

1. Name of the registry: Prospero.

2. Unique Identifying number or registration ID: CRD42020167327.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked): https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=167327.

Guarantors

Michiel C. Warlé and Gabby T.J.A. Reijnders-Boerboom.

Data availability statement

All extracted data can be made available upon reasonable request by contacting the corresponding author. Data collection forms and analytic code were based on the PRISMA guidelines and can also be shared upon reasonable request.

Supplementary Material

js9-109-1400-s001.docx ^{(634.7KB, docx)}