Randomized clinical trials

Fawzy A, Balbaa MA, Hagag M.

BJS Open 2023; 7: zrac173.

There was no difference in the rate of patient satisfaction between these two procedures in the 78 patients studied. Most patients in both groups considered the result excellent.

Comment: Both these new techniques are effective.

Schrag D, Uno H, Rosovsky R, Rutherford C, Sanfilippo K, Villano JL et al, for the CANVAS Investigators.

JAMA 2023; 329: 1924-1933.

This study included 671 patients with cancer and a new episode of VTE. After six months, the risk of recurrent VTE was not inferior with oral anticoagulation (6.1per cent), compared to those receiving low molecular weight heparin (8.8 per cent). Bleeding complication rates were similar.

Comment: Reassuring, and easier for patients.

Maemoto R, Noda H, Ichida K, Miyakura Y, Kakizawa N, Machida E et al.

Ann Surg 2023; 277: 727-733.

Lavage with povidone iodine before skin closure for 1 min was compared to saline lavage in this study that included 941 procedures. The rate of surgical site infection was similar: 7.6 per cent with iodine and 5.1 per cent with saline, P=0.154.

Comment: the fact that the wound is flushed may be more important than what with.

Patel SP, Othus M, Chen Y, Wright GP, Yost KJ, Hyngstrom JR et al.

N Engl J Med 2023; 388: 813-823.

Some 313 patients with advanced melanoma undergoing surgical resection followed by one year of adjuvant Pembrolizumab, were randomly allocated three additional doses of Pembrolizumab preoperatively. After two years, event-free survival was significantly better with the neoadjuvant treatment: 72 versus 49 per cent, P=0.004.

Comment: This is important evidence for the benefit of neoadjuvant treatment for melanoma.

Kimball AB, Jemec GBE, Alavi A, Reguiai Z, Gottlieb AB, Bechara FG et al.

Lancet 2023; 401: 747-761.

This analysis of two identical trials included over one thousand patients. Secukinumab given fortnightly was more effective in both trials at reducing the signs of symptoms of hidradenitis than placebo (or Secukinumab every four weeks).

Comment: Effective, with good safety profile.

The TRACE Study Investigators.

Ann Surg 2023; 277: 375-380.

Over 5000 patients were included in this multicentre trial. Thirty-day mortality was 0.56 per cent  (n = 14) in the control and 0.44 per cent (n = 12) in the intervention group (odds ratio 0.74, 95% Confidence interval 0.34–1.62).

Comment: Would results have been different if the focus had been on high risk patients only?

Jain A, Greig AVH, Jones A, Cooper C, Davies L, Greshon A for the NINJA Collaborative.

Br J Surg 2023, 110: 432-438.

In this study that included 451 children, discarding the nail after repair was not associated with an inferior outcome (infection/cosmesis), but reduced overall costs by a mean of 84 euros.

Comment: The nail can be discarded.

Almeida I, Gonqalves AC, Corrêa FB, Castro J, Guirro E, Farina JA et al.

Ann Surg 2023; 277: 198-205.

Patients with burn contractures at least one year old were randomized to split skin grafting alone, or with the addition of one of three dermal matrix preparations. Detailed scar assessment showed no improvement over skin grafting alone with any of the dermal matrices.

Comment: Disappointing outcome for the dermal matrices.

Wei Y, Wu J, Yuxin C, Fan K, Yu X, Li X et al, on behalf of the PL-5 Investigators.

Ann Surg 2023; 277: 43-49.

This study examined the use of a new antimicrobial spray compared to silver sulphadiazine in 220 patients with a skin wound infection. Use of the spray at four varying doses all improved healing rates significantly (P<0.05).

Comment: This seems to be working more like an antiseptic.

Sali L, Ventura L, Mascalchi M, Falchini M, Mallardi M, Carozzi F et al.

Lancet Gastro Hepatol 2022; 7: 1016-1023.

Some 15,000 people were randomized and invited. Attendance was lower for CT colonography (26.7 per cent versus 33.4 per cent who participated in all three faecal testing rounds). Although the detection rate was higher with CT colonography (5.2 versus 3.1 per cent, P=0.0002). Overall detection (referral to work-up) was higher after three rounds of faecal testing (7.5 versus 2.7 per cent, P<0.0001).

Comment. Interesting that compliance with screening may be more important than the accuracy of the test.

Albers KI, Polat F, Helder L, Panhuizen IF, Snoeck MMJ, Polle S et al, for the RECOVER Study Collaborators.

Ann Surg 2022; 276: e664-e673.

A total of 178 patients were randomized to high or low pressure pneumoperitoneum. Quality of recovery score was significantly improved on day one after low pressure pneumoperitoneum (167 versus 159, P=0.005). Inflammatory markers were also reduced. Pain scores were lower and there were fewer infectious complications.

Comment: But is the view as good?

Abbassi F, Muller SA, Steffen T, Schmied BM, Warschkow R, Beutner U et al.

Br J Surg 2022; 109: 1216–1223.

Sixty patients were randomized to one of two caffeine doses given three times a day, or a control group. There was no measurable effect from caffeine; time to first bowel movement was not improved: 68.2 h after caffeine versus 67.3 h in controls (P=0.887).

Comment; Well there goes the excuse to ask for a coffee after your operation.

Kang B, Ibrahim S, Weil A, Reynolds K, Johnson T, Wilson S et al.

Ann Surg 2022; 276: 975-980.

Patients who had skin surgery with primary wound closure were randomly allocated to one of two different laser therapies for eight weeks or a control group. Fifty-two patients were eligible for analysis. Laser treatment improved wound score after 36 weeks, particularly with regards to scar thickness, stiffness and erythema.

Comment: Whether laser treatment improves wound healing still remains unproven.

NIHR Global Research Health Unit on Global Surgery.

Lancet 2022; 400: 1767-1776.

In this study that included over 13,000 procedures, this simple change reduced the rate of surgical site infection from 18.9 to 16 per cent, P=0.0032.

Comment: I really like studies that appear simple and generalisable. Hospitals around the world with the highest infection rates stand to see the biggest differences.

Tandon S, Ensor ND, Pacilli M, Laird AJ, Bortagaray JI, Stunden RJ et al.

Br J Surg 2022; 109: 1087-1095.

After six weeks, cosmetic scores were best with sutures and worst with glue (P=0.014), but the differences were no longer apparent in this study that included 295 children (333 surgical wounds).

Comment: No strong conclusion about which method should be used.

Sevinç B, Damburaci N, Karahan O.

Dis Colon Rectum 2022; 65:735-741.

Two hundred patients had their pilonidal sinus curetted and then treated with either plasma gel or phenol crystals. Healing was more likely after a single application of plasma – 96 versus 53 per cent; it was also quicker (6 versus 10 days, P<0.001) and recurrence after one year was lower (4 versus 12 per cent).

Comment: Extraordinary result that needs confirmation to ensure it is generalisable.

Moncrieff MD, Bastiaannet E, Underwood B, Francken AB, Garioch J, Damude S et al.

Ann Surg 2022; 276: e208-e216.

Some 388 patients were enrolled into a study comparing standard follow-up with a reduced frequency schedule. There was no difference in recurrence rates between the groups. Three quarters of patients in both groups found their recurrence by self-examination.

Comment: Routine follow-up may not help patients with melanoma; perhaps an approach teaching them self-examination would be better.

Livingstone E, Zimmer L, Hassel JC, Fluck M, Eigentler TK, Loquai C et al, on behalf of the Dermatologic Cooperative Oncology Group.

Lancet 2022; 10358: 1117-1129.

Some 175 patients were randomized into one of three treatment arms. After a median of 49.2 months, recurrence-free survival was 64.2 per cent after the combination of nivolumab and ipilimumab, 31.4 per cent after nivolumab alone and 15 per cent after placebo; P<0.0001. Combination treatment also improved overall survival.

Comment; Combination treatment should be standard in this group.

Krige A, Brearley SG, Mateus C, Carlson GL, Lane S.

BJS Open 2022; 6: zrac055.

Pain scores were improved after epidural analgesia for the first 24h (pain score 33 versus 50.5; P=0.018) in this study that included 131 procedures. After 72h, pain control was better with a rectus sheath catheter (4.5 versus 12.5; P=0.019), which was also more cost effective.

Comment: Perhaps the compromise is opiate analgesia to supplement the rectus sheath catheter.

Fortelny RH, Adrade D, Schirren M, Baumann P, Riedl S, Reisensohn C et al.

Br J Surg 2022; 109: 839–845.

This study included 425 patients who had a midline laparotomy. After one year, the rate of ultrasound-detected incisional hernia was lower after short stitch (5 to 8mm bites every 5mm) than long stitch repair (10mm bites every 10mm): 3.3 versus 6.4 per cent, P=0.173.

Comment: The risk of incisional hernia was low with both methods.

Lastunen KS, Leppäniemi AS, Mentula PJ.

Br J Surg 2022; 109: 588-594.

Some 185 Patients with suspected appendicitis were randomised to imaging with ultrasound followed by CT if needed, or observation. More patients in the imaging group had treatment for appendicitis (72 versus 57 per cent; difference 15, confidence interval 1 to 29 per cent). Rates of complicated appendicitis and negative appendicectomy were similar.

Comment: Some patients with appendicitis do not need surgery; I think we knew that, but this is good science.

Smith S, Ridley S, Gani, J, Carroll, R, Lott, N, Hampton, et al.

Ann Surg 2022; 275: 842-848.

In this study that included 3123 procedures, surgical site infection rates were similar with all skin preparations: povidone iodine with alcohol 10.9 per cent; chlorhexidine with alcohol 11.1 per cent;  aqueous povidone iodine 12.6 per cent. There were no adverse events or differences in secondary outcomes.

Comment: This challenges the perceived benefits of alcoholic skin preparation.

Gao X, Liu Y, Zhang L, Zhou D, Tian F, Gao T et al.

JAMA Surg 2022; 157: 384-393.

The study included 230 patients undergoing major abdominal surgery. Early (three days after surgery) feeding reduced the rate of nosocomial infection from 18.4 to 8.7 per cent (P=0.04) compared with late feeding (eight days after surgery).

Comment: The findings in many studies are consistent: early feeding is beneficial.

Sessler DI, Pei L, Li K, Cui S, Chan MTV, Huang Y et al, on behalf of the PROTECT Investigators.

Lancet 2022; 399: 1799-1808.

Of the 5013 patients studied, those who had warming had a higher core temperature at the end of the surgery (37.1 versus 35.6 degrees C). Warming did not reduce the rate of complications: 9.9 versus 9.6 per cent, respectively, P=0.69.

Comment: This does not confirm previous studies of benefits from heating, maybe because the unheated patients did not get too cold.

Devereaux PJ, Marucci M, Painter TW,  Conen D, Lomivorotov V, Sessler DI et al, for the POISE-3 Investigators.

N Engl J Med 2022; 386: 1986-1997.

A total of 9535 procedures were studied. Tranexamic acid reduced bleeding complications (11.7 versus 9.1 per cent, P<0.001), but slightly increased the risk of cardiovascular complications (14.2 versus 13.9 per cent, P=0.04).

Comment: Is the trade-off worth it, except for specific procedures with high blood loss?

Luke JL, Rutkowski P, Queirolo P, Del Vecchio M, Mackiewicz J, Chiarion-Sileni V et al on behalf of the KEYNOTE-716 Investigators.

Lancet 2022; 399: 1718-1729.

Adjuvant treatment with pembrolizumab every three weeks for up to 17 cycles was examined in the study that included 976 patients with median follow-up of 20.9 months. Treatment improved recurrence-free survival: 85 versus 76%, hazard ratio 0.61, 95 per cent confidence interval 0.45 to 0.82.

Comment: This difference is valuable, but again, it remains to be seen how late outcomes are affected.

Lenet T, Baker L, Park L, Vered M, Zahrai A, Shorr R et al.

Ann Surg 2022; 275: 456-466.

Fourteen trials including 8641 procedures were studied. A restrictive blood transfusion policy did reduce the volume of blood transfused without increasing postoperative morbidity. Two of the trials did show adverse composite outcomes for restrictive transfusion.

Comment: The result isn’t definitive, but further evidence that restrictive transfusion is generally not harmful.

Hemmingson O, Binnermark F, Odensten C, Rutegard M, Franklin KA, Haapamaki MM.

BJS Open 2022; zrac007.

The study was terminated at planned interim analysis when 116 patients were included. Healing was significantly better after Karydakis flap (14 versus 49 days, P<001). Recurrence rates were similar.

Comment: This study has a clear result, but there remain a number of alternatives to the Karydakis flap.

Seidel D, Lefering R.
Ann Surg 2022; 275: e290-e298.

This study included 539 patients with an abdominal wound. Negative pressure dressings reduced hospital stay (11.8 versus 13.9 days P=0.047), but duration of actual treatment within 42 days longer: 22.8 versus 30.6h, P<0.001.

Comment: Not as good as a formal cost effectiveness analysis, which is sorely missing for this technology.

Deeken F, Sanchez A, Rapp MA, Denkinger M, Brefka S, Spank J et al, for the PAWEL Study Group.
JAMA Surg 2022; 157: e216370.

A total of 1470 patents were randomly allocated to a prevention programme or standard care. Overall, the programme reduced the rate of postoperative delirium: 5.3 per cent versus 6.9 per cent in controls, P=0.047. It was ineffective after cardiac surgery.

Comment: Small effect only.

Li T, Li J, Yuan L, Wu J, Jiang C, Daniels J et al, for the RAGA Study Investigators.
JAMA 2022; 327: 50-58.

Some 950 patients over 65 years old were included. Postoperative delirium within seven days was recorded in 5.1 per cent in those who had a GA, compared to 6.2 per cent who had spinal or epidural, P=0.48.

Comment: surprisingly low rate; reproducible in other specialties?

Bechara FG, Podda M, Prens EP, Horvath B, Giamarellos-Bourboulis EJ, Alavi A et al.
JAMA Surg 2021; 156: 1001-1009.

The effect of perioperative adalimumab continued for ten weeks after radical surgery (wide excision followed by secondary intention healing) was examined in this study that included 206 patients. A good clinical response was more common with adalimumab therapy: 48 versus 34 per cent, P=0.049.

Comment: Some possible hope for patients with a difficult disease to manage.

Lancet 2021; 398: 1687-1699

This 2 x 2 factorial trial studied two different antiseptics for skin preparation (alcoholic chlorhexidine and aqueous iodine) and the use of antiseptic (triclosan) coated sutures among 5788 patients undergoing surgery. The overall surgical site infection rate was 22 per cent. There was no benefit from using chlorhexidine or triclosan sutures over the cheaper alternatives.

Comment: Another example of more expensive interventions that have no proven value in low and middle income countries. Why does that appear to be different in higher income countries?