SS Stapled vs Excision Hemorrhoidectomy Long-term Results of a Prospective Randomized Trial Franc H. Hetzer, MD; Nicolas Demartines, MD; Alexander E. Handschin, MD; Pierre-Alain Clavien, MD, PhD Hypothesis: stapled hemorshoidectomy offers several advantages over excision hemorthoidectomy, including reduced postoperative pain, a reduced hospital stay, and an earlier recovery time. Furthermore, stapled hemor- rhoidectomy is associated with lower hemorrhoidal re: ‘currence on long-term follow-up. Design: A randomized prospective trial, Patients were blinded to the operation technique used. Follow-up oc- ‘curred at 1 and 3 weeks and 12 months postoperatively Setting: A university hospital providing primary, sec~ Patlents: Forty patients with second. and third-degree hemorrhoid disease were randomized to undergo either stapled or excision hemorthoidectomy. Two patients were excluded. all patients were subject to a follow-up ex Interventions: Stapled hemorrhoidectomy (Longo technique) vs excision hemorthoidectomy (Ferguson technique). Main Outcome Measures: Operating time, postop- erative pain (measured by the visual analog seale), hos- pital stay, histologic features, morbidity, defecation habit, continence, recovery time (Feturn to work), and hemor- shoid recurrence at 1 year. Results: Siapled vs excision hemorthoidectomy was as- sociated with a significantly reduced operating time (30 scores (visual analog score) on the first + postoperative days (day 1: 2.7 vs 63; day 2: 1.7 vs 6.3; day 3: 0.8 vs 5.4; and day 4:0.5 vs 48, where 0 indicates no pain, and 10, maximum pain; P.001), and an earlier return to work (6.7 vs 20.7 days; P=.001). There were no differences for stapled vs excision hemorrhoidectomy in length of hos- pital stay (24 ys 2.1 days), complications (3 [15%] of 20 patients vs 5 [25%] of 20 patients), and recurrence rate (1 [5%] of 20 patients vs 1 [5%] of 20 patients) Conelusions: Stapled hemorthoidectomy is associated with reduced postoperative pain, earlier recovery time and return to work, and a similar recurrence rate com- pared with the excision technique. Provided further clini cal trials confirm these findings, stapled hemorthoidec- tomy may become a future gold standard, Arch Surg, 2002;137:337-340 XCISION hemorthoidectomy fs associated with signifi- cant postoperative pain be: ‘cause of trauma ofthe sens tive anal mucosa (ano- compared with conventional hemorthotd- ectomy. Faster wound healing and less postoperative pain have also been ob- served.* Larger stidies®® comparing the 2 .chniques confirmed less postoperative From the Departmen for Visceral and Transplantation Surgery, University Hospital of Zurich, Zurich, Switzerland (aepnuyreo) ARCHSURGNOLTA WARIO derm). Furthermore, the patients have to maintain a precise wound dressing to pre- vent local infection, because local wound exposure may lead to fecal contamination and prolonged wound healing." In a small series including 23 patients, it was shown that the stapled hemorrhoidectomy ini- tilly described by Longo’ leads to less post- ‘operative pain, a shorter postoperative hos- pital stay, and shorter recovery time in patients with third-degree hemorrhoids (©2002 American Med pain and an earlier return to work in the stapler group but showed no difference in total hospital stay and overall complica- tions, However, severe complications fol- lowing stapled hemorthoidectomy have been reported in 0.12% ofthe cases; these complications include sphincter lesions, persistent postoperative pain, rectal per” forations, and even lethal sepsis." These controversial results suggest the need for further prospective evalus- 1 Association, All rights reserved.PATIENTS AND METHODS Between January 1, 1999, and July 31, 2000, 42 patients ‘wth symptom second-orhid-depfcehemorod dis ase, according to the grading of Milles" were included in ths prospective randomized study. Two patients re fused to patcpate, The study was approved by the local cthiescomnitee. Following writen informed consent the patients were allocated by drawing lote—generatedran- Eornization to unvdergo either stapled hemorrhoidectomny (dhe Long technique?) {n=20) or excision hemorrhoid tctomy (dhe Ferguson technique") (4-20) During the hos Pill stay, the patients were not informed ofthe technique performed, bu this information was given duringa -week follow-up examination on request Beease the patios were blinded othe technique used the same cae and dressing othe anal region was performed in both groupe inthe post operative period “The operation was performed under citer general an- cstheis (in 22 [33% ofthe 40 patient) of spl anes thesia (in 18 [45% ofthe 4 patents), depending onthe silent: preference andthe atesthesiologet's alice, and rth sain surgeon (FH.H1), who was experienced inco- lorecal and proctologic surgery, with a previous learning carve in stapled hemorshotdectony (>30 procedures) Par ttents were placed in a positon for ithotomy. A cleaning tema was given preoperatively. Before extubation, the pat tents received baste analgesia inteavenously (2g of ace $nminophen) patente who were operated om using spinal amcshesiareeelved 0.5% bupivacaine hydrochloride lo- cally No antibiotics were given in this tra. The hemor- Thokdetomy nthe conventional group was performed a- Cording ta the Ferguson technique. The base of the hemorthold was excised andthe wound was sutured with 25.0 poyglactino 0 thread (Vil rpc Ethicon, Ine, NoF- derstedt, Germany), In 13 patients, this was a pile exci slon;in 7, itwasa 2plleeXcsion Inthe stapler group a crularanadiatator (CAD3: Ethicon Endo Surgery ne, Norderstedt) was introduced toreduce the prolapse ofthe anoderm and parts ofthe anal tnucous membrane. Alter removal of the obturator, the pro- Inpsed mucous membrane fle into the amen ofthe cr cll anal distor. Thus, a purse-string suture, nonab- Sorbable surgical of 2-0 polypropylene Prolene; Ethicon, Inc) was placed circumferemally 3 10 5 cm above the dentate line trough the window ofthe purs-ting suture smoscape (PSA 35; Ethicon Endo-Surgery, Inc). Subse- Gquentlya hemorrhoidal circular stapler (4333, Ethicon Eco Surgery, nc wis positioned an fied. Final inboth soups alerosaicendoanal dressing pongo anal Fer Fosan, Socborg, Denmarl) was applied. The operating time tvs defined asthe time from he beginning of the operation Unt the application ofthe endoanal desing Allpaents rceveda normal diet postoperatively and were ven lactulose for preventing hafd stool. Patents in both groupe were equested to perform the same cleaning oftheanal eglon 2103 tes per day using a shower. The sane typeof external anal dressing was applied, and the pallene agreed not to inspect the anal region themselves {omainan blinding during the postoperative period "Apainscore datasheet (visu and scale) wae filled cut by the patients postoperatively (0 indicates no pn nd 1, maximum pat), Pat scores were evaluated 12 hours Inter and on the next 3 consecutive postoperative days by 4 surgeon not involved inthe operation, Prin therapy con- Sted of basic analgesia (acetaminophen) and addition of subcutaneous infetions of meperaine hydrochloride, 35 mg every 3106 hours, on request, At dcharge from the hospital, he patente received lcilose, 20 mi dally, sd bate analgeta (acetaminophen) All specimens were analyzed histologically alter he- smatoryircosn staining to delet skeletal or smock muscle fibers. A continence score was evaluated using the Wil Hime score" preoperatively and after 12 weeks "A follow-up examination was performed 3 and 12 weeks postoperatively by an independent surgeon (not & tnember of the operating tenn). Endosonographic contol orsphincter manometry were only performed if clinical ev dence of sphincter lesions was presen. Postoperalivecom- plications (vith special regard to rectal stenosis), defec- tion babi, frequency, and return to work postoperaivel twere erated. In addition, + I-yea follow-up examin don was performed with special regard to hemorrhoid recurrence At this examination, defecation habits were Cvaluated and s procioogic examination was performed. Salistical analyse was performed by the Mann Whitney testand the Wilcoxon ran suny test for un- ited data, P05 was regarded a significant A power Ealetlation was not perforin before the study. tion anda randomized comparison of stapled vs conven- tional hemorrhoidectomy. To our knowledge, there are few data comparing the long-term results of stapled vs excision hemorrhoidectomy.”" This prospective random- od study analyzes the outcome of stapled vs excision hemorrhoidectomy in patients with second- oF third- degree hemorrhoid disease blinded to the operation tech- nique used, with special regard to the long-term results and recurrence rate ——___ m7 Forty patients were operated on for second-degree (n=12) or third-degree (n=28) hemorrhoids, accord- ing to the Milles classification."? Ten patients with second-degree hemorrhoids were treated previously by rubber band ligature, and 2 refused the rubber band. (aepnuyreD) ARCHSORGNOLTA WR (©2002 American Med These 12 patients were operated on and included in the study Patient characteristics were comparable for age, sex, and grade of hemorthoid disease. The characteristics of the patients in the 2 groups are as follows: characte stapled etsion ooo croup Total No. 0 20 Doge of moras ‘Second 6 Tire 4 Pecutnt hemor 2 Age, mean ange). sna,a2.72) Nale-femal to 155 The overall operating time was 30 minutes (range, 15-45 minutes) in the stapler group and 43 minutes (cange, 25-60 minutes) in the excision group (P<.001). 1 Association, All rights reserved.nrc oe oS acon | 7 Eo a L ay Posoperave pan evalsted by he visual analog scale. Data ar given as rmeanes0, Surgery was performed under general anesthesia in 12 fof the 20 patients in the stapler group and in 10 of the 20 patients in the excision group, while spinal anesthe- sia was applied in 8 and 10 patients, respectively Using the visual analog seal, mean pain scores were 2.7 (range, 0-8), 1.7 (range, 0-6), 0.8 (range, 0-3), and (0.5 (range, 0-2) on days 1, 2, 3, and 4, respectively, in the stapler group; in the excision group, the respective values were 6.3 (range, 0-10),6.3 (range, 1-10),5.4 (range, 1-9), and 4.8 (range, 1-10). The average amount of pain in the stapler group was significantly lower than in the excision group (P=.001) (Figure) The mean length of the hospital stay after hemor- rhoidectomy was 2-4 days (range, 1-4 days) in the sta- pler group and 2.1 days (range, 1-# days) in the exci- sion group; this difference was not statistically significant (P2.17). Patients returned to work at an average of 6.7 days (range, 2-14 days) inthe stapler group and 20.7 days. (eange, 7-45 days) in the excision group (P=.001). Histologic examinations of resected specimens fe vealed small parts of skeletal muscle fibers in 3 patients (25%), all in the excision group (P=.43). Smooth muscle fibers were found in 4 patients (20%) in the stapler group and in 5 patients (25%) in the excision group (P=.80) ‘Of the 40 study patients, perioperative complica- tions observed included bleeding in 2 patients and peri- anal thrombosis in | patient in the stapler group, and uri nary retention in 1 patient and suture dehiscence in 4 patients in the excision group, all occurring within the first postoperative week The 2 bleeding complications occurred within 2 hours following surgery and required subsequent operation. In ‘one patient, a bleeding arterial vessel had to be sutured; i the other patient the bleeding stopped after internal com- pression with a balloon catheter for 30 minutes, The total postoperative complication rate was 15% (3 of 20 pa- tients) in the stapler group and 25% (5 of 20 patients) in the excision group (P=.60). There were no deaths in el- ther group: and at 1 year, recurrent hemorrhoidal disease ‘occurred in 1 (5%) ofthe patients in both groups. ‘A follow-up examination after 3 and 12 weeks (fol- low-up, 100%) revealed impaired wound healing be- ‘cause of suture dehiscence in 4 of the 40 patients, all in the excision group. No impaited wound healing was ob- served in the stapler group. No cases of incontinence were ‘observed during the follow-up period. The Williams score, ‘evaluating for incontinence, was 1.0 preoperatively and, (aepnnyeo) ARCHSORGNOLTA WARIO (©2002 American Med postoperatively in the stapler group and 1.1 preopers- tively and postoperatively in the excision group. ‘Aer 1 year, a total of2 patients presented with second- degree recurrent hemorrhoidal disease; one was operated on by the excision technique and one was operated on by stapled hemorrhoidectomy. Both recurrent hemorthoids were treated successfully with a rubber band ligature After I year, there were neither signs of rectal ste- nosis nor peritectal fistulas in the patients had residual perianal pain, Because neither ther group, and none of signs of sphincter damage nor incontinence were ob- served in both groups, we did not perform postopers- tive endosonography or a manometric examination. —_ kml _ The use of a stapler in the treatment of hemorrhoids r mains controversial. The results of a prospective ran- domized study comparing the gold standard—excision hemorrhoidectomy—with the new stapler technique, with patients blinded to the type of procedure, are impor- tant, We observed a significant reduction of postopers- Live pain in the patients who underwent stapled hemor- shoidectomy. Pour patientsin the stapler group were pain free on the first operative day. Our results confirm those of 5 previous randomized trials!" on stapled vs con- ventional hemorrhoidectomy. The total operating ime was significantly shorter with the stapler technique in this trial (30 vs 43 minutes; P<001). However, this time s 5 to 10 minutes longer than observed by others“? We routinely performed a dilata- tion of the anal sphincter before stapler introduction, which ‘may explain this slightly prolonged operating time. Except for one postoperative bleeding episode, which required a blood transfusion and revision, no other vere complications were observed in the stapled hemor- rhoidectomy group, especially no local or systemic in- fections.” The bleeding observed resulted most likely from aan undetected vessel within the stapler line. This com- plication may be prevented ifadequate hemostasis around the stapler line is obtained routinely and each bleeding vessel is sutured. concern about stapled hemorthoidectomy isthe po- tential risk of strictures after rectal wall resection."® Even though total of 20 hisiologic examinations of stapled hem- sled partial rectum wall margins with smooth e was no clinical sigh of rectal strictures or ste- nosis aller 12 months, It was not possible to differentiate the smooth muscle fibers originated [rom the rectal wall or from the internal sphincter. However, we observed no {incontinence in any patient atany time. The histologie analy- sis revealed skeletal bers in 3 patients in the excision group but only ina small quantity. This finding excludes the pos- sibility that the excision hemorrhoidectomay may have been performed too deeply. Moreover, there was no incont rence in any patient at any time. ‘As expected in the excision group, there were always parts of muicos from the anal eanal. Ifthe resection is high enough above the dentate line in the stapler group, ab- sence of mucosa ofthe anal canal in the histologic finding. may exclude potential sphincter damage. Squamous cell epithelia were never demonstrated, but in patients, small 1 Association, All rights reserved.parts of epithelium of the anal canal were observed. One ‘of these patients complained about more postoperative pain (visual analog scale score: day 1, 7; day 2, 6; and day 3, 3) than the mean of patients in the stapler group. This find- ing is most likely a result of resection margin close to the sensitive epithelium of the anal canal area, and may ex- plain the bad results on pain observed in a famous study ‘on stapled hemorthoidectomy performed in an estab- lished colorectal center by Cheetham etal’ The research- cers decided to interrupt their study because patients in the stapler group had significantly more postoperative pain. This publication remained controversial, and was challenged in several letters to the editor.”"* In our series including a 1-y up, none of the patients had persistent residual pain. This suggests that great attention has to be given to the level ofthe sta- pler line 3 t0 5 em above the dentate line, as initially ree- ‘ommended by Longo" and confirmed by others" There- alter, the complete resection line is located above the anodermal line, out of the sensitive nerves, thus explain- ing the absence of pain The pathophysiologic background of the treatment of hemorrhoidal disease by stapler isdifferent than the patho- physiologic basis for excision hemorrhoidectomy, and is |eing controversially discussed, The complete circular mu ‘cosa cranial to the hemorrhoidal plexus is resected, allow- ing reduction of mucosa prolapse by mucosa lifting and by fixing the prolapsed mucosa at the rectum wall. The redue- Lion of arterial blood flow tothe hemorrhoidal plexusisprob- ably not the main point of the treatment. A Doppler inves- ligation with preoperative and postoperative measurement ‘ofthe arterial inflow to the hemorrhoidal cushion did not show any significant differences.** The repositioning ofthe prolapsed mucosaand thereby the improvement ofthe ve nous reflux may be the key of the treatment, but furtherin- vestigations are necessary to claily Uhis point The indication for stapled hemorrhoidectomy in our study included third-degree hemorthoids and second- degree hemorrhoids alter an unsuccessful nonoperative treatment (eg, a rubber band ligature). Fourth-degree hemorrhoids are usually not regarded asa contraindica- tion for stapled hemorthoidectomy,” but were not in- cluded in the present study. The incidence of hemorrhoid recurrence did not dif- fer in the 2 groups within the L-year follow-up, but a longer follow-up should be observed. We did not evaluate the postoperative analgesic med ‘ation taken, which may limit the interpretation of this ob- servation; however, even though both patient groups had free access to minor analgesics, the stapler group had sig- nificantly less pain than the excision group (average vi- sual analog seale score, 1 4s 5.7; P<.001). Ina random- ized tral, Mehigan et al? reported a 50% reduction in postoperative pain medication consumption i patients who lunderwent stapled hemorrhoidectomy compared with pa- tients who underwent excision (Milligan- Morgan) hem- ‘orthoidectomy. The rather small number of patients who ‘were included in the study limits the interpretation of the results. However, the results of our prospective, random- ized, patient-blinded study are in accordance with the re sults of the 4 other studies** comparing stapled hemor- thoidectomy with excision hemorthotdectomy. rear Follo (aepnnyeo) ARCHSORGNOLTA, WARIO (©2002 American Med We conclude that with an adequate handle of the stapler procedure, including sufficient sphincter dilatation be- fore stapler placement, a resection line at least 3 cm above the dentate line, and a cautious hemostasis during sur- gery, stapled hemorrhoidectomy isasafe and reliable pro- cedure inthe treatment of second. and third-degree hem- orrhoids. It offers a similar elinical outcome as excision hemorrhoidectomy while offering a significantly shorter op- erating time, significantly reduced postoperative pain, an earlier retur to work, and low recurrence at 1 year. Pro- Vided Further clinical rials confirm the hndings of our study, stapled hemorrhoidectomy may become new standard in the treatment of hemorrhoid disease This study was presented atthe 88th Annual Mecting of the Swiss Society of Surgery, Luzern, Switzerland, June 7, 2001 Corresponding author and reprints: Nicolas Demar- tines, MD, Department or Visceral and Transplantation Sur- _gery, University Hospital of Zurich, Ramistrasse 100, 8091 ‘Zurich, Switzerland (e-mail nicolas demartines@chi.usz.ch), —_ I mm” 1. Hot Seo Chen, Tan Leong AFP. Radar coals wal top nd sed haemo. B JS. 14784172070. 2. Rowse M, Ble, Herangnsy OM. Creare naosectoy (sald asmorhiscomy) versus convertor! Neemarthadecomy:encomaed tents Laney 200036877078 8 Longo A Teatnent of beara: dies by reduson of mucosa and hen ‘ol propos wh crear sutuig even prea, Pre Ing af te ot Wend Congas of Eneeoe Surges. Balog, fa Mon ie Pubic Co 1008777 7a 4 Raf O Bir aD Randi Si rl of strc re tpl flsednasnorodeony BI Surg 200087 1350-185. 5 Metgan Moroso Harty J Sapling proces forhaemorais e- ‘us lig Mergen heortotcony: random oa a Lancet Soo. 70078. 6. Ho Goong WK Tsang ea Supednanarsdeomy-—cast and fetes: ranomind,colé ral nding eananc cag. 0 rectal mame, nd adoual aon aecemer sp vee mons 1 1678, IM, itm PO, Karr A, iil RAS. Parse gen) sr Spe hamriany.Lancet 200, 8 HoH Teng, Tang Lyn D Eu Ki, Seou-Chn Foal spine "es am taping eure dues neal: dome, contol ‘uy wth fea esound and anorectal manoney. eCalon stam, Sodas eo173 ©. Mar Mc Mectrcalnemarieony winger staples: waring toes Tagua in Fee. Sie Sug 1000818115. 10. Harald Ks! Complisons ater stapladhemoroiectomy re fut af2n guy in Germany, Cloclgy 20028816 1. Gano Atomare OF, Gab Faia G, CoS Prospective anda ules ral arin sled with open hasmerhaecomy. Sr Su, oavasseoere lc Ober atone pen ines, Surg nace On 018.28 407-56 Fergron Jason JA Gosed morhastony. Dis Gln Rect 18% eer Villars Ms, Pal 4 Geog 80, Developme ofan cloticay sims neo al sper Lancet 1081388116611, Haloy RE, KgsareD Lie heigl spsis te taped ham ‘hodctomy fe] Lancet 2000385 810 Fas il Ey poner of staplng esque o:5 760-769. Pesci Pana slanted tr ane 20002562188 Harlech Paster ped sorely eter). Lace 200, seater Ga CB, Malcolm AL Pan ater stapled haemarhoetory foe Lancet oo'62180, Jogi Bock Pan ater sped harmonica [ay] anes 2000 seater, 21, Longo Panatr sapere lt. Lane 2000256:2180- io 2. Kolber GW, a Fain oh cee of arridetany th Lon fel eu by Oop tasound othe ae sal sper beta) Protea 201-7 2 i 4 16 16 sortideony. Lancet " ie 0 0 1 Association, All rights reserved.