Page 37 - Read Online

Bongiolatti et al. Mini-invasive Surg 2020;4:41 I http://dx.doi.org/10.20517/2574-1225.2020.28 Page 5 of 11

Table 2. Study type, year of publication and main characteristics of the included studies
Author Type of esophagectomy Conversions EBL (mL) Type of anastomosis
Boone et al. [18] TT MKE 7 (15%) 625 Cervical handsewn end-to-side
Puntambekar et al. [27] TT MKE 0 80 NA
Dunn et al. [19] TH 5 (12.5%) 97.2 Cervical mechanical end-to-end
Sarkaria et al. [21] TT ILE+MKE 10 (48%) 307 cm 3 Mechanical circular endo-to-end (ILE)
Cervical handsewn end-to-side (MKE)
Suda et al. [32] TT MKE NA 144 Cervical handsewn end-to-side or
cervical handsewn end-to-end
de la Fuente et al. [36] TT ILE NA 146 NA
Yerokun et al. [39] NA 28 (12.1%) NA NA
Weksler et al. [17] NA 6.7% NA NA
van der Sluis et al. [14] TT MKE 3 (5%) 120 Cervical handsewn end to side
Harbison et al. [16] TT 11 (11%) NA NA
Yang et al. [22] TT MKE 2 (0.7%) 211 Cervical mechanical end-to-end
Tagkalos et al. [28] TT ILE NA NA Cervical mechanical end to side
Sarkaria et al. [31] TT ILE+MKE NA 250 NA
Yun et al. [29] TT MKE+ILE 3 (2.3%) 110 Mechanical circular
EBL: estimated blood loss; TT: trans-thoracic; MKE: McKeown esophagectomy; ILE: Ivor-Lewis esophagectomy; NA: not available; TH:
trans-hiatal

are summarized in Table 1. No formal statistical procedure (meta-analysis) was performed. One study was
[31]
[14]
a RCT , and the other 13 were observational studies including one prospective and 12 retrospective
studies published from 2009 to 2019. In addition, four papers had propensity-matched analysis [22,28,29,39] and
three were multi-center studies [16,17,39] .

RESULTS
Three-hundred and twenty studies were initially identified from the electronic databases and after screening
and reviewing, 14 were included for final analysis. Table 2 shows the main characteristics of the included
studies.

Intra and post-operative outcomes
Conversion rates were reported in ten papers and were much different from the early experience to the
latest study [Table 2]. The largest multi-center studies, published in 2016 and 2017, showed a conversion
rate ranging from 6.7% to 12.1%; in the RCT, the rate is lower (5%), probably due to the large experience
gained by the Dutch group . Operative time is significantly longer for RAMIE in comparison with open
[14]
esophagectomy (OE) [14,29,31] and MIE [16,22,29] .

[19]
Dunn et al. in 2012 demonstrated the feasibility of the trans-hiatal approach in a cohort of 40 patients
with 2.5% mortality at 30 days, but there was quite a high incidence of overall post-operative complications:
anastomotic leaks without the need for re-operation (n = 10, 25%); recurrent laryngeal nerve injuries (n =
14, 35%) and pneumonia (n = 8, 20%) [Table 3]. The use of this approach has gradually decreased in favor
of trans-thoracic esophagectomy because the lymph node dissection is more extensive with trans-thoracic
esophagectomy and more accurate surgical and pathological staging could be obtained. Trans-hiatal MIE
or RAMIE could be useful approaches in patients with severe lung function impairment or other relevant
co-morbid conditions because one-lung ventilation and thoracic incisions are not required .
[9]
The overall 90-day post-operative mortality rate after trans-thoracic esophagectomy was reported in
ten papers and ranged between 0% to 9% without any difference between two or three field esophagecto
my [14,16-19,22,28,29,31] . The RCT published by van der Sluis et al. reported comparable in-hospital mortality
[14]
rates between patients who underwent RAMIE (2%) and OE (4%) (P = 0.62). The 90-day mortality rate was
[16]
not significantly higher for RAMIE patients (2% vs. 9%; P = 0.11). Multicenter analysis by Harbison et al.