World J Surg (2009) 33:1795–1801 DOI 10.1007/s00268-009-0106-3
Healing of Stoma Orifices: Multicenter, Prospective, Randomized Study Comparing Calcium Alginate Mesh and Polyvidone Iodine Mesh Olivier Brehant Æ Patrick Pessaux Æ Nicolas Regenet Æ Jean Jacques Tuech Æ Fabrizio Panaro Æ Georges Mantion Æ Vincent Tassetti Æ Paul Antoine Lehur Æ Jean Pierre Arnaud
Published online: 15 July 2009 Socie´te´ Internationale de Chirurgie 2009
Abstract Background The aim of this multicenter, randomized, prospective study was to reveal a difference in terms of a guided healing period in the case of stoma orifices after reestablishing digestive continuity by comparing an alginate mesh with a polyvidone iodine mesh. Methods Between April 2004 and September 2005, a total of 73 patients were randomized into two groups: A (alginate mesh) and M (polyvidone iodine mesh). The groups were comparable for demographic data, indications for and the type of stoma, and perioperative data. The main evaluation criterion was percentage healing at the 28th postoperative day (D28); and the secondary criteria were healing time, rate of infectious complications, and number of dressing changes applied. Results The mean percentage healing at D28 was 91% in group A and 87% in group M (p = 0.49). The mean healing time was 31 days in group A and 32 days in group M (p = 0.80). One parietal abscess (3%) occurred in group
O. Brehant P. Pessaux F. Panaro J. P. Arnaud (&) Digestive Surgery Department, University Hospital, 4 rue Larrey, Angers 49000, France e-mail:
[email protected] N. Regenet P. A. Lehur Digestive Surgery Department, University Hospital, Place Ricordeau, Nantes, France J. J. Tuech V. Tassetti Digestive Surgery Department, Hospital Centre, Mulhouse, France G. Mantion Digestive Surgery Department, University Hospital, Besancon, France
A (p = 0.37). The mean number of meshes used was 13 ± 5 in group A and 18 ± 8 in group M (p \ 0.005). Conclusion The use of an alginate mesh for guided healing of stoma orifices after reestablishing digestive continuity allows effective healing within a normal period of time with a lower number of meshes.
Introduction Every year in France, more than 30,000 new digestive stomas are carried out. Almost half of them are temporary [1], in which case they allow the acute phase of an infectious, inflammatory, or neoplastic pathology to pass or they protect an anastomosis carrying a risk of fistula. When reestablishing digestive continuity (RC), the residual skin opening requires special treatment. There are two documented approaches: primary closure (PC) at the same time as the operation or secondary closure (SC) with guided healing of the stoma (SO) using dressings. PC consists of suturing the wound planes with or without drainage. SC consists of suturing the aponeurotic planes of the abdominal wall associated with guided healing of the subcutaneous plane and the skin. Over the last three decades, few studies in the literature [2–11] have documented test results allowing a comparison to be made between these two closure techniques. Retrospective studies [2–10] have revealed the superiority of the SC technique. The rate of infection of the SO is lower than that for PC, where it can reach 40% [9]. Only one randomized prospective study [11] has been documented in the literature, which means that no conclusion can be reached on the superiority of one or the other of these techniques. One disadvantage of SC is the long period of guided healing, which may exceed 30 days [12].
123
1796
The aim of this multicenter, prospective, randomized study was to compare the healing period of the two types of dressings used: a calcium alginate (Algosteril) mesh (Laboratoires Brothier, Nanterre, France) dressing or a dressing impregnated with polyvidone iodine for guided healing of the SO.
World J Surg (2009) 33:1795–1801
comparing both mean values with a standard deviation of 25% (risk a of 5%, risk b of 10%, D 20%). The secondary judgement criteria were the time for complete healing of the SO and the rate of infectious complications. An economic evaluation involved calculating the direct costs per patient. In each group, the total cost of the equipment and the cost of nursing care, multiplied by the number of dressings per patient, was calculated.
Materials and methods Procedure Patients All patients who were candidates for RC through SO closure between April 2004 and September 2005 were considered eligible for the present study. Exclusion criteria were age 17 years or younger, emergency surgery, true and serious allergy to iodine or to one of the treatments used, a stoma affecting a digestive segment other than the small intestine or the colon, patient with dysthyroidia, patients requiring additional protection (pregnant or breast-feeding woman, subject under wardship or guardianship), and informed consent not signed. The protocol was approved by the local ethical committee (Comite´ Consultatif de Protection des Personnes dans la Recherche Biome´dicale, Angers, France). Informed consent was obtained from each patient before they entered the study. Study design and randomization The main criterion for evaluation was the percentage filling of the SO cavity at day 28 (D28). This percentage was calculated by comparing the volumes of the SO at D0 and D28. The volume of the SO cavity was measured in accordance with a standardized protocol to limit any measurement bias. After cleaning the wound, a transparent film of the Tegaderm type 3 M (Laboratoires 3 M Sante, Cergy-Pontoise, France) was placed over the SO. The volume of the cavity was measured by the quantity of 0.9% NaCl solution, injected with a syringe via the transparent film, required to fill the whole cavity. According to our center’s experience, the mean duration of healing for stoma openings after reestablishing digestive continuity was 41.9 days. On D28, 50% of the residual cavity was filled. Assuming a healing rate with the reference treatment (mesh impregnated with polyvidone iodine) of 50% at D28, to demonstrate a 20% difference (i.e., healing of 70% at D28 with Algosteril) it was necessary to include 64 patients. This calculation was carried out using a bilateral test with a risk a of 5% and a risk b of 10% (i.e., a power of 90%). We enrolled 70 patients (i.e., 35 patients in each group) to take into account the possibility of 10% being lost to follow-up. This calculation is based on
123
No digestive preparation was carried out prior to the intervention. All patients received antibiotic prophylaxis with cefazolin, and a daily subcutaneous injection of lowmolecular-weight heparin sodium as deep venous thrombosis prophylaxis. The type of anastomosis, manual or mechanical, was left to the choice of the surgeon. With regard to closing the SO, the two aponeurotic layers of the sheath of the rectus abdominis muscles of the abdomen were sutured. The volume of the residual cavity was evaluated in accordance with the established protocol to obtain a reference value for the patient. The product, designated by randomization, was then placed in the SO. The size of the alginate mesh was 5 9 40 cm. The width of the mesh impregnated with polyvidone iodine was 2 cm. The length of the dressings was based on the volume of the whole cavity. No other local drainage was used. Dressings were changed every 2 days, with only one dressing whenever possible. Data collection, crossover, and postoperative follow-up The patients were examined every day while they were in hospital. The volumetric measurement of the SO was carried out every 7 days for 1 month on days 0, 7, 14, 21, and 28. All the patients were reviewed during the sixth postoperative week for the end-of-study visit. Definitions Complications were stratified in accordance with Clavien’s classification [13]. • • • • •
Grade I: complications inducing any deviation from the normal postoperative course Grade II: complications requiring pharmacologic treatment Grade III: complications requiring surgical, endoscopic, or radiologic intervention Grade IV: life-threatening complications requiring intermediate or intensive care unit management Grade V: complications resulting in the death of the patient
World J Surg (2009) 33:1795–1801
1797
Surgical site infection (SSI) has been defined according to the Centers for Disease Control and Prevention (CDC) classification: infections of the incision scar at the surgical site or parietal infectious complications [superficial (infections of skin and subcutaneous tissues) or deep infections (extending to fasciae and/or muscles)], deep organ infection, interorgan spaces, and cavity infections or deep infectious complications [14].
Assessed for eligibility (n=76)
Refused to participate (n=3)
Excluded (n=2) No RC
Randomized (n=71)
Statistical analysis Data were expressed as mean ± standard deviation (SD). Statistical analysis was performed using SPSS software (version 10.0; SPSS, Chicago, IL, USA). Categorical variables were analyzed by the chi-squared test or Fischer’s exact test as appropriate. Continuous variables were analyzed by Student’s t-test or the Mann-Whitney test as appropriate; p \ 0.05 was considered significant. All statistical analyses were two-tailed and based on the intentionto-treat concept.
Group A (n=37)
Refused to participate (n=2)
Excluded (n=3) fistule Analysed (n=37)
Results Between April 2004 and September 2005, a total of 76 patients were eligible for this trial. Three patients refused to give their consent, and in two patients digestive continuity was not reestablished during the intervention. A total of 71 patients were randomized: 45 men (63%) and 26 women (37%) with a mean age of 57 ± 19 years. Distribution was 37 patients in the alginate mesh group (group A) and 34 patients in the group for the mesh impregnated with polyvidone iodine (group M). Five patients (all in group M) were excluded from the analysis either because they refused to follow the protocol (n = 2), or because of the occurrence of an anastomotic fistula exteriorized through the old stoma orifice (n = 3) (Fig. 1). The statistical analysis therefore involved 66 patients: 41 men (62%) and 25 women (38%) with a mean age 58 ± 19 years. There were 37 in group A and 29 in group M. Stoma indications were colon cancer in 23 patients (35%), complicated diverticulitis in 15 patients (23%), or another etiology (Crohn’s disease, ulcerohemorrhagic rectocolitis, pancreatitis, mesenteric infarctus, trauma, hemorrhage) in 28 patients (42%). There were 30 colostomies (45%) and 36 ileostomies (55%), terminal (42%) or lateral (58%). A protective stoma was placed in 27 patients (41%). The mean period between the creation and the closure of the stoma was 195 ± 192 days (range 6–1251 days). Reestablishment of digestive continuity was carried out electively incision (n = 35) or by median laparotomy (n = 31). There was no difference between the two groups in terms of preoperative and perioperative data (Table 1).
Group M (n=34)
Analysed (n=29)
Fig. 1 Study design
Table 1 Patient data Parameter
Group A (n = 37)
Group M (n = 29)
p NS
Age
60.4 ± 18.3
55.5 ± 19.7
Sex (M/F)
24/13
17/12
NS
ASA
1.92 ± 0.6
2.0 ± 0.5
NS
Colostoma
17
13
NS
Ileostoma
20
16
NS
Type of stoma
ASA American Society of Anesthesiologists
Overall morbidity was 15.2% (n = 10), with no difference between group A (16.2%) and group M (13.8%) (p = NS). The number of of SSIs did not differ between the two groups: 5.4% (n = 2) for group A and 5.1% (n = 2) for group M (p = 0.8). Mortality was zero in each group (Table 2). The mean percentage healing for the 66 patients was 89% ± 21% at D28. This was calculated from the volumes recorded for each group (Table 3). The percentage healing at D28 was 91% for group A and 87% for group M (p = 0.49) (Table 4, Fig. 2). The mean time for SO healing after RC was 31 ± 8 days for group A and 32 ± 7 days for group M (p = 0.80) (p = NS). On the D45, a total of 59 patients (89%) had healed, 34 in group A and 25 in group M (p = NS).
123
1798
World J Surg (2009) 33:1795–1801 100
Table 2 Morbidity Condition
Alginate dressing (n = 37)
Polyvidone iodine dressing (n = 29)
p
Grade II
6
3
0.59
Surgical-site infection
3
2
Wound infection
1
0
Deep infection
2
2
Ileus
1
0
Urinary infection Pneumopathy
1 1
0 1
Grade III
1
1
Evisceration
0
1
Rectovaginal fistula
1
0
ALGO IODO
50
0 J14
0.99
J21
J28
Fig. 2 Development of percentage healing compared with D7 depending on the product used. ALGO alginate mesh, IODO polyvidone iodine mesh
Table 5 Calculation of costs Cost parameter Table 3 Mean volumes (in milliliters) depending on the product used
Price of equipment/ dressing (euros)
Time of sampling
Group A (n = 37)
Group M (n = 29)
p
Price of nurse/ dressing (euros)
Day 0
8.09
6.62
0.21
Mean price/ patient (euros)
Day 7
8.30
6.21
0.24
Day 14
4.89
3.50
0.30
Day 21
1.87
2.08
0.80
Day 28
0.83
1.43
0.39
Day 45
0.04
0.38
0.27
Table 4 Mean percentage healing depending on the product used Interval (days)
Group A (n = 37)
0–7
–2.30a
4.27
0.64
0–14
42.48
50.91
0.38
0–21
79.31
76.64
0.69
0–28
90.91
87.19
0.49
7–14
46.02
41.97
0.62
7–21
79.61
73.30
0.37
7–28
90.82
86.40
0.47
a
Group M (n = 29)
p
The negative sign signifies an increase, not a decrease, in volume
The mean number of dressing changes used to achieve total healing was lower in group A (13.2 ± 5 dressings) than in group M (17.9 ± 8 dressings) (p \ 0.005). The lower number of dressing changes used in group A is likely to reduce costs for total healing. The mean cost was 268.0 euros per patient in group A and 286.5 euros per patient in group M (Table 5). In total, the use of an alginate mesh in SO healing resulted in a saving of 18.50 euros over the total healing time.
123
Group A
Group M
6.7
2.4
13.6
13.6
268
286.5
Discussion This series represents the first prospective randomized study comparing the efficacy of two types of dressing change in healing SOs after RC. The results show that, for similar efficacy, calcium alginate is more economical without increasing morbidity or mortality. This advantage is significant because of the number of stomas created each year. In 1991, a presentation to the Surgical Society of Lyon [1] noted that more than 30,000 new stomas were established per annum in France, including all types, 49% of which were temporary. Current indications for stoma include a large number of pathologies—inflammatory, infectious, neoplastic. In our study, the distribution was approximately 45% colostomies and 55% ileostomies. The usual indications for stoma are represented in our series but with a significant proportion (approximately 43%) for pathologies (excluding cancer and diverticulitis) that are grouped together as inflammatory, infectious, and even trauma. It is difficult to relate these data to those in literature, given, for example, the high percentage of stomas established as a result of traumatic lesions of the colon or rectum described by some North American teams [15]. The period of time prior to RC varies from one surgical team to another. Most publications concern RC using Hartmann’s procedure. Closure of the colostoma usually takes place 4 months after it is created [16]. After this period of time, operating problems seem to be reduced
World J Surg (2009) 33:1795–1801
1799
[17]. Some authors [18, 19] even go so far as to recommend a preclosure period of at least 6 months to reduce morbidity, especially the occurrence of an anastomotic fistula. The data in our study also suggest the advantage of a fairly long period. The mean period prior to closure of stomas for the 66 patients was 196 days (i.e., slightly more than 6 months). We report a rate of anastomotic fistula formation of 4.1% (n = 3). Two patients presented a clinical fistula and one patient a radiological fistula noted in the presence of a perianastomotic abscess, confirmed by computed tomography (CT), with rectal opacification. Analysis of literature identified 10 studies over the last three decades relating to the treatment of SO after RC. Secondary closure (SC) consists of closing various aponeurotic planes, both subcutaneous and cutaneous, with or without drainage. It offers the advantage of immediately closing the opening, which does not require any prolonged dressing, but at the price of more significant local infectious morbidity than if the orifice is left open. SC consists of suturing the aponeurotic planes of the abdominal wall and then guided healing of the subcutaneous level and the skin. It allows a reduction in local infectious morbidity with less abscess formation but at the price of a longer healing time.
Of the 10 studies reported in literature, 9 are retrospective [2–9] (Table 6). The assessment criteria were mortality and local and general morbidity, in particular the levels of infection of the surgical wound, an abdominal hernia associated with the SO, and anastomotic fistulas. They revealed the superiority of the SC technique. The level of infection of the SO was lower than that for a PC. However, none of these trials provided quantitative data about the healing time according to the type of closure used. For PC, we can find data according to whether subcutaneous drains were present. For SC, no study compared two or more types of dressing. Only one randomized prospective study [11], published in 1985, evaluated the level of infectious complications of the SO according to the type of parietal closure. In this study, 105 patients, 97% of whom had stomas that were put in place because of traumatic lesions, were randomized into three arms: (1) PC (n = 38), (2) PC with two subcutaneous drains (n = 29), and (3) SC (n = 38). The patients were prepared for surgery by combining mechanical digestive preparation with antibioprophylaxis. Digestive preparation began 2 days prior to surgery with a liquid diet. Three evacuation enemas were administered: the first the day before hospital admission and the others in the morning and evening prior to surgery. A nasogastric tube was inserted the day before the operation for irrigation of 6 L of
Table 6 Retrospective studies comparing primary closure and secondary closure Study
Year
Patients
Bell [2]
1974
284 (12 years)
Todd [3]
1979
206
Closure
Mortality
Morbidity
1.8%
Infection
Conclusion
20.8%
SC [ PC (NS)
PC 27%
SC [ PC (NS)
SC 5.5% Garber [4]
1982
80 (6 years)
PC 41 SC 39
Salley [5]
1983
166 (8 years)
PC 135
26%
14% PC 17%
SC [ PC (NS)
SC 10% 0
2.4%
1.2%
PC feasible if antibiotics
PC 32% SC 26%
SC [ PC except if antibiotics/Flagyl
15%
SC [ PC
PC 19%
p = 0.009
SC 31 Perry [6]
1983
105 (6 years)
PC 74
2%
SC 31 Pittman [7]
1985
126 (10 years)
PC 42
0
33%
SC 84
SC 2.8% Hubens [8]
1987
79 (8 years)
PC 44
2.5%
43%
19%
Hackam [9]
1995
95 (6 years)
SC [ PC p \ 0.05
SC 35 PC 54
29.5%
SC [ PC
SC 41
PC 41%
p \ 0.05
SC 15% Riesener [10]
1997
548 (22 years)
PC 279 SC 269
2%
48.2%
35.4%
Antibioth. ? Colon prep.
PC primary closure, SC secondary closure
123
1800
Ringer’s lactate solution. A catheter was placed in the stoma orifice for irrigation with 1 L of 0.9% NaCl solution containing antibiotic biotherapy. This biotherapy was also administered per os in three doses the day before the intervention as well as at the preoperative and postoperative stages. The results showed a 4.8% rate of infection of the stoma opening (n = 5; one in group 1, one in group 2, three in group 3). These differences were not significant (p = 0.32). Berne et al. [11] concluded that they could not recommend SC in the treatment of SO particularly in the case of careful mechanical digestive preparation and antibioprophylaxis. However, for the PC group, the difficulty implementing the protocol for patient preparation limited the scope of the results. This is all the more applicable because at the moment the recommendations are not to apply colon preparation [20]. In the light of these different results, we have opted for secondary closure of stoma openings. During the course of our study, the level of infection in the closure of colostoma openings was low. One single abscess occurred in the 66 patients (1.5%). The patient was from group A, but the difference was not significant (p = 0.37). The products that can be used for healing a wound are a dressing that is plain or impregnated with antiseptic, hydrocolloid, alginate, hydrocellular agent, a hydrogel or carbon dressing. The use of local antiseptics as prophylaxis or to heal already infected wounds is widely accepted nowadays [21–23]. The mesh impregnated with polyvidone iodine is a gauze dressing impregnated with an antiseptic from the group of iodinated derivatives. It is used for its healing properties thanks to its fibrillary structure and for its antiseptic properties obtained from polyvidone iodine. Algosteril is a hemostatic healing alginate that reduces the risk of infection; it is indicated in wounds in which there is peri- and/or postoperative bleeding, for acute or chronic loss of substances, infected wounds, epistaxis, and rhinosinus surgery. Algosteril allows activation of hemostasis, activation of macrophages, fixation of bacteria, effective drainage, and retention and optimization of the tissue repair process. A randomized multicenter clinical trial published in 1998 [24, 25] compared the advantage of Algosteril mesh with a mesh impregnated with polyvidone iodine in the local treatment of a pilonidal sinus with abscess. A group of 70 men between 18 and 30 years of age presenting with a postoperative infected pilonidal sinus were randomized into two arms: Algosteril mesh and mesh impregnated with polyvidone iodine. The study showed that the cavity filled and healed in a larger number of patients in the alginate group. On the 21st day of dressing application, the reduction in volume of the cavity was 67.1% with alginate and 45.6% with the dressing impregnated with polyvidone
123
World J Surg (2009) 33:1795–1801
iodine. This difference is statistically significant (p = 0.0007). The authors concluded that the action of alginate on the healing process results in more rapid closure of infected wounds than is seen with conventional treatments and in patient acceptability, particularly because of the reduced pain. In this study, the pilonidal sinus was used as an experimental model of an infected wound. From an economic point of view, the results are interesting insofar as a reduced healing period and the need for fewer dressings in favor of Algosteril allow a reduction in indirect costs, such as home nursing care and absence from work. After the first few days, we noted that the Algosteril dressings took more than 48 hours to become saturated, allowing treatment to be spread out.
Conclusions This randomized prospective multicenter study comparing two medical devices in post-RC healing is the first of this type to be carried out. The percentage healing at day 28 and the healing time are the same as with a reference system (mesh impregnated with polyvidone iodine). However, the need for fewer new mesh applications in the alginate group resulted in reduced costs. For closure of a stoma after RC, we therefore recommend secondary closure with guided healing using an alginate.
References 1. Berard P, Verges J, Conge M et al (1991) French stomatherapy faced with 30,000 new stomas per annum. Lyon Chir 87:505–508 2. Bell GA (1974) Closure of colostoma: a review. Am J Proctol 25:77–84 3. Todd GJ, Kutcher LM, Markowitz AM (1979) Factors influencing the complications of colostoma closure. Am J Surg 137:749– 751 4. Garber HI, Morris DM, Eisenstat TE et al (1982) Factors influencing the morbidity of colostoma closure. Dis Colon Rectum 25:464–470 5. Salley RK, Bucher RM, Rodning CB (1983) Colostoma closure: morbidity reduction employing a semi-standardized protocol. Dis Colon Rectum 26:319–322 6. Perry GJ, Payne JE, Chapuis PH et al (1983) Complications in colostoma closure. J R Coll Surg Edinb 28:174–177 7. Pittman DM, Smith LE (1985) Complications of colostoma closure. Dis Colon Rectum 28:836–843 8. Hubens G, Minten L, Hubens A et al (1987) Colostoma closure: still a hazardous procedure. Acta Chir Belg 87:205–210 9. Hackam DJ, Rotstein OD (1995) Wound infection during stoma closure. Can J Surg 38:191 10. Riesener KP, Lehnen W, Hofer M et al (1997) Morbidity of ileostoma and colostoma closure: impact of surgical technique and perioperative treatment. World J Surg 21:103–108 11. Berne TV, Griffith CN, Hill J et al (1978) Colostoma wound closure. Arch Surg 120:957–959
World J Surg (2009) 33:1795–1801 12. Smit R, Walt AJ (1978) The morbidity and cost of the temporary colostoma. Dis Colon Rectum 21:558–561 13. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240: 205–213 14. Centers for Disease Control (1992) CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Cont Hosp Epidemiol 13:606–608 15. Thal ER, Yeary EC (1980) Morbidity of colostoma closure following colon trauma. J Trauma 20:287–291 16. Roe AM, Prabhu S, Ali A et al (1991) Reversal of Hartmann’s procedure: timing and operative technique. Br J Surg 78:1167–1170 17. Keck JO, Collopy BT, Pjet R et al (194) Reversal of Hartmann’s procedure: effect of timing and technique on ease and safety. Dis Colon Rectum 37:243–248 18. Mosdell DM, Doberneck RC (1991) Morbidity and mortality of ostoma closure. Am J Surg 162:633–636 19. Pearce NW, Scott SD, Karran SJ (1992) Timing and method of reversal of Hartmann’s procedure. Br J Surg 79:839–841
1801 20. Slim K, Vicaut E, Panis Y et al (2004) Meta-analysis of randomized clinical trials of colorectal surgery with or without mechanical bowel preparation. Br J Surg 91:1125–1130 21. Lagoutte F, Fosse T, Jasinski M (1992) Iodized polyvidone (betadine) and prevention of postoperative infection: a multicenter study. J Fr Ophtalmol 15:14–18 22. Molloy RG, Brady MP (1993) A modified technique of delayed primary closure using a povidone iodine wick: influence on wound healing in an experimental model. Ir J Med Sci 162: 297–300 23. Regnard JF, Libert JM, Rojas Miranda A et al (1989) Antibiotics and antiseptic prophylaxis in thoracic surgery: controlled study. Pathol Biol 37:477–480 24. Guillotreau J, Andre J, Flandrin P et al (1996) Calcium alginate and povidone iodine packs in the management of infected postoperative wounds: results of a randomized study. Br J Surg 83: 861 25. Deligny M (1998) Advantage of Algosteril dressing and a dressing impregnated with polyvidone iodine in the local treatment of pilonidal sinus with abscess. Wound Healing 12:231–233
123
