Professional Documents
Culture Documents
Described in the literature dehiscence rate in the adult population is 0.3-3.5%, and in the elderly group
as much as 10%. In about 20-45% evisceration becomes a significant risk factor of death in the periop-
erative period.
The aim of the study was to identify the main risk factors for abdominal wound dehiscence in the
adult population.
Material and methods. The study included patients treated in the 3rd Department of General Surgery,
Jagiellonian University Collegium Medicum in Cracow in the period from January 2008 to December
2011, in which at that time laparotomy was performed and was complicated by wound dehiscence in
the postoperative period. For each person in a research group, 3-4 control patient were selected. Selec-
tion criteria were corresponding age (± 2-3 years), gender, underlying disease and type of surgery
performed.
Results. In 56 patients (2.9%) dehiscence occurred in the postoperative period with 25% mortality.
The group consisted of 37 men and 19 women with the mean age of 66.8 ± 12.6 years. Univariate
analysis showed that chronic steroids use, surgical site infection, anastomotic dehiscence/fistula in the
postoperative period and damage to the gastrointestinal tract are statistically significant risk factors
for dehiscence. Two first of these factors occurred to be independent risk factors in the multivariate
analysis. In addition, due to the selection criteria, a group of risk factors should also include male
gender, emergency operation, midline laparotomy, colorectal syrgery and elderly age (> 65 years).
Logistic regression analysis did not show that a particular surgeon, time of surgery or a particular
month (including holiday months) were statistically significant risk factor for dehiscence.
Conclusions. Wound dehiscence is a serious complication with relatively small incidence but also
high mortality. Preoperative identification of risk factors allows for a more informed consent before
patient’s treatment and to take measures to prevent or minimize the consequences of complication
associated with it.
Key words: dehiscence, risk factors
Spontaneous durability of the wound in the rity. If the healing process is disturbed, this
first day after surgery is virtually non-existent can lead to partial or complete dehiscence of
and gradually increases with time. In third individual layers of the sutured wound or to
week after surgery the durability equals 20% wound dehiscence along its entire depth, called
of the initial strength, and after 6-12 weeks it eventration. Usually dehiscence or eventration
reaches 70-80% (1). Sutures placed during the occur 4 to 14 days after surgery (on the 8th day
surgery should allow the tissues the necessary on average). The incidence of this laparotomy
time to regain structural and functional integ- complication is estimated to be 0.3–3.5%, and
Unauthenticated
Download Date | 5/12/19 10:07 AM
566 J. Kenig et al.
Table 1. Indications and procedure scope in the group of patients with wound dehiscence
Unauthenticated
Download Date | 5/12/19 10:07 AM
Risk factors for wound dehiscence after laparotomy – clinical control trial 567
analysis of risk factors already described in non-parametric tests. For all variables, both
papers, i.e.: age (>65 years), gender (male), qualitative and quantitative, a univariate re-
neoplastic disease, COPD or pO2 <60 mm Hg gression analysis was performed. Variables
and pCO2 <30 mm Hg, malnutrition (decreased demonstrating statistical significance in the
albumin level <30 g/l or decreased body weight univariate analysis were included in a multi-
>=10%), presence of sepsis, obesity ≥30 kg/m2, variate analysis to determine independent risk
anemia (hemoglobin level <10 mg/dl), diabetes, factors. The null hypothesis (H0) was rejected
hypertensive disease, coronary heart disease, at the established level alpha = 0.05.
chronic steroid treatment in the last 12 months
and tobacco smoking. Factors related to sur-
gery included the operating surgeon as a risk Results
factor as well as the setting of surgery (elective
or emergency surgery), exact time of surgery, During the study period 1,879 laparotomies
procedure type, technique, suture type used were performed. Postoperative wound dehis-
for closure of the layers and additional use of cence was observed in 56 patients, that is in
anti-eventration sutures. Postoperative com- 2.9% of procedures. The group consisted of 37
plications were analyzed in detail, with par- men and 19 women, which means there was a
ticular emphasis put on infection of the surgi- statistically significantly greater number of
cal site. A comparison was performed of dehis- men in this population (p < 0.05). The mean
cence time, length of hospital stay, necessity age was 66.8±12.6 years. Postoperative wound
of treatment in the intensive care unit (includ- dehiscence occurred on average after 9.8±6.5
ing length of stay in the ICU) and number of days (median: 8 days). Mortality in this group
deaths. reached 25%. There was a statistically sig-
nificantly greater number of patients admitted
in an emergency setting: 45 (80.4%) vs. 11
Statistical analysis patients admitted in an elective setting
(p<0.01). No statistically significant age differ-
To describe the study results qualitative ence was reported in patients undergoing
and quantitative data were used. Quantitative elective surgeries. On the other hand, in an
parameters were expressed as the mean value emergency setting women were statistically
± standard deviation. The remaining cases significantly older than men (74.2±10.7 vs.
were coded using Arabic numerals. The data 63.4±13.1 years; p < 0.05).
were analyzed using STATISTICA 10.0 soft-
ware suite (StatSoft). The Shapiro-Wilk W test
and the Kolmogorov-Smirnov test with the Comparison of female and male populations
Lilliefors correction were used to verify the with dehiscence
normality of distribution of results. Depending
on the result of the normality of distribution, A comparison of the study group members
the data were analyzed using parametric or based on gender (tab. 2) did not reveal any
Unauthenticated
Download Date | 5/12/19 10:07 AM
568 J. Kenig et al.
Surgery type:
– stomach/duodenum 3 (15,8%) 2 (5,4%) 0,2
– gall-bladder 1 (5,3%) 4 (10,8%) 0,5
– small intestine 1 (5,3%) 8 (21,6%) 0,2
– large intestine 11 (57,9%) 17 (46%) 0,6
– other 3 (15,8%) 6 (16,2%) 0,9
Time to wound dehiscence 9,45+6,9 10,5+5,9 0,56
Vertical incision 18 (95%) 30 (81%) 0,16
Stay in the ICU 8 (42,1%) 18 (48,6%) 0,64
Biochemical tests:
– WBC 13,6±5,6 12±6,2 0,3
– HCT 34,7±5,5 36,6±8 0,4
– HGB 10,8±1,7 12,8±5,4 0,15
– CRP 152,2±130,9 111,3±92,5 0,3
– albuminy / albumin 25±4,2 33,3±13,4 0,05
– białka / protein 49,4±7,9 53,6±15,6 0,5
– kreatynina / creatinine 96,3±28,5 100,8±52,7 0,4
Death 6 (31,6%) 8 (21,6%) 0,41
difference in the setting of surgery, presence/ and male populations with dehiscence revealed
lack of neoplastic disease and other concomi- no statistically significant difference (p = 0.16).
tant diseases, time to dehiscence, tobacco
smoking, chronic steroid treatment, number
of previous surgeries, type of the present sur- Comparison of results in the study and
gery, length of hospital stay (including length control groups
of stay in the ICU), leukocyte count, hematocrit
level, hemoglobin level and protein level. Sta- A comparison of the study group with the
tistically significantly lower albumin level at selected control group demonstrated that the
admission was reported in the female popula- study group was characterized by a statisti-
tion compared with the male population cally significantly greater incidence of surgical
(33.3±13.4 vs. 25±4.2 g/l; p < 0.05), with no site infection and greater incidence of circula-
difference in leukocyte count and CRP level. tory failure in the postoperative period. The
On the other hand, opening of the abdomen patients also required a longer stay in the ICU,
using a vertical incision was statistically sig- and their hospital stay was statistically sig-
nificantly more frequent in the group with de- nificantly longer (tab. 3). The remaining pa-
hiscence than other types of opening (48 vs. 8 rameters listed in tab. 3 showed no statisti-
persons; p < 0.01). A comparison of the female cally significant difference.
Table 3. Comparison of the study group with the selected control group
Unauthenticated
Download Date | 5/12/19 10:07 AM
Risk factors for wound dehiscence after laparotomy – clinical control trial 569
Grupa rozejścia
Grupa kontrolna /
Czynnik / Factor rany / Wound P value
Control group
dehiscence
Procedures involving opening of the gastrointestinal tract 42 (76,4%) 122 (72,6%) 0,16
Neoplastic disease 24 (43,6%) 71 (42,3%) 0,85
Concomitant diseases:
– hypertensive disease 28 (50,9%) 92 (54,8%) 0,61
– coronary heart disease 23 (41,8%) 61 (36,3%) 0,46
– diabetes 8 (14,5%) 28 (16,7%) 0,71
– COPD 9 (16,4%) 16 (9,5%) 0,16
– other 42 (76,4%) 116 (69%) 0,3
BMI:<20,5/20,5-30/>30 kg/m2 (n) 12/31/13 24/119/24 0,12
Previous surgery (n) 25 (45,5%) 91 (54,2%) 0,26
Tobacco smoking 15 (27,3%) 41 (24,4%) 0,67
Chronic steroid treatment 7 (12,7%) 6 (3,6%) 0,84
Surgical site infection 34 (61,8%) 23 (13,7%) <0,01
Anastomotic dehiscence/fistula 5 (8,9%) 5 (3%) 1
Circulatory failure (n) 21 (37,5%) 43 (25,6%) 0,04
Vertical incision (n) 47 (83,9%) 143 (85,1%) 0,95
Anti-eventration sutures (n) 13 (23,2%) 27 (16,1%) 0,2
Procedure time (n):
7-15.00 24 (42,9%) 81 (48,2%) 0,68
15.01-23.59 26 (46,4%) 73 (43,5%) 0,73
00.00-6.59 6 (10,7%) 14 (8,3%) 0,89
Stay in the ICU (n) 25 (45,5%) 39 (23,2%) 0,001
Length of hospital stay (days) 38,3±27,1 15,8±12,9 <0,01
Biochemical tests:
– WBC 12,5±6,0 11,6±5,9 0,3
– HCT 35,9±7,2 37,7±6,4 0,1
– HGB 12,2±4,6 12,4±2,2 0,7
– CRP 127,13±108,2 111,49±100,27 0,5
– albumin 30,2±12,9 32,69±9,6 0,3
– protein 52,44±13,9 56,85±13,8 0,2
– creatinine 104,4±62,5 91,98±63,7 0,3
Time to wound dehiscence (days) 9,8±6,5 -
Death 13 (23,2%) 34 (20,2%) 0,6
Table 4. Analysis of risk factors in the study group using logistic regression
Unauthenticated
Download Date | 5/12/19 10:07 AM
570 J. Kenig et al.
Multivariate analysis demonstrated that group with dehiscence in our study, we found
only chronic steroid treatment and surgical statistically significantly greater number of
site infection are independent risk factors for men than women (37 vs. 19 – ratio of 2:1). In
postoperative wound dehiscence. one of the published studies, it was attributed
to tobacco smoking, which in most cases was
seen among men. In our study, tobacco smok-
Discussion ing did not constitute a risk factor, and men
did not smoke more frequently than women.
Wound dehiscence is a serious postoperative Frequency comparison of factors predisposing
complication associated with high mortality to wound dehiscence (tab. 2) also demonstrat-
reaching 45%. Incidence reported in the lit- ed no statistically significant difference be-
erature is between 0.3 and 3.5%, but there are tween men and women, except for a lower al-
individual reports of incidence as high as 10%. bumin serum level in the female population.
In our study, wound dehiscence occurred in Another explanation for this might be, re-
2.9% of the cases, and mortality in this popula- ported in certain papers, increased pressure
tion was 25%, which did not significantly differ inside the abdominal cavity generated by men,
from results found in the literature (8, 9, 10). which translates into greater forces exerted on
the main wound (11). On the other hand,
other studies associate the greater incidence
Patient-related factors of dehiscence with lower collagen production
in the wound in the male population, which is
Most researchers agree that the male gen- probably associated with the effect of estrogens
der is a risk factor. Through analyzing the (12).
Unauthenticated
Download Date | 5/12/19 10:07 AM
Risk factors for wound dehiscence after laparotomy – clinical control trial 571
In most studies, age is also a risk factor accordance with reports in published papers.
(13-17). In our study, logistic regression anal- It stems from numerous factors, such as a
ysis showed no such relation, but, as has been general worse condition (including worse nutri-
mentioned earlier, in line with the assump- tion) and greater risk of surgical field con-
tions of this study, the control group members tamination in case of patients undergoing
were selected in accordance with the patients’ emergency surgery. Some studies raise also
age. While analyzing patients in the study the issue of the time when such surgeries take
group, one can easily notice a significant over- place, which is often at night, potentially lead-
representation of people over 65 years old (only ing to suboptimal management of the abdom-
12 persons (21%) were younger than 65). inal wall layers (17, 21, 22). A comparison of
Chronic steroid treatment is also a fre- elective surgery and emergency groups par-
quently reported risk factor for wound dehis- ticipating in our study demonstrated no sta-
cence. It is believed that these drugs inhibit tistically significant difference between ana-
healing of the original wound and delay gran- lyzed factors, except for a statistically signifi-
ulation tissue formation, and their chronic use cantly lower albumin level and total protein
increases the number of intra- and postopera- level as well as a higher CRP serum level,
tive complications, which is associated, among which is associated with severity of disease
other things, with suppression of the hypotha- being the cause of admission in an emergency
lamic-pituitary-adrenal axis (18). In our study, setting. An interesting observation is the fact
multivariate regression analysis showed that that among patients with wound dehiscence
chronic steroid treatment is an independent admitted in an emergency setting there were
risk factor for wound dehiscence. statistically significantly fewer people with an
It has been demonstrated that tobacco oncological disease, while among patients with
smoking results in abnormal wound healing wound dehiscence admitted in an elective set-
due to increased tissue hypoxia, impairment ting there were statistically significantly more
of neutrophil-killing mechanisms, decreased people with an oncological disease (p < 0.01).
collagen production and disturbed ratio of This might be associated with the fact that
proteases and their inhibitors in the wound most oncological patients admitted in an emer-
itself, which may predispose to wound dehis- gency setting underwent a resection with
cence. In our study, however, chronic tobacco creation of a stoma (a procedure of shorter
smoking was not found to be a risk factor for duration and associated with fewer periopera-
wound dehiscence. Both the study and the tive complications compared with resection
control groups had a similar percentage of with one-step restoration of the gastrointesti-
habitual smokers (12). nal tract continuity).
Interestingly, diabetes, anemia, neoplastic
disease, previous laparotomy, obesity and un-
derweight were not identified as risk factors in Treatment-related factors
our study. Vascular changes resulting from
hypertensive disease, micro- and macroangiopa- Logistic regression analysis did not show
thy associated with diabetes or an overall bad that a particular surgeon (also resident vs.
condition caused by neoplastic disease should specialist), the time of surgery or a particular
predispose to wound dehiscence. Mixed opinions month (including holiday months) were statis-
are also seen in the literature (17, 19, 20). tically significant risk factors for dehiscence.
As far as we are concerned, it is the first mono-
graph analyzing the above factors in such
Disease-related factors detail. The literature features also papers in-
dicating a higher risk of wound dehiscence in
Emergency surgery is another risk factor case of opening the fascia by means of electro-
for wound dehiscence identified in most stud- coagulation, due to formation of marginal
ies. While analyzing the group with wound necrosis. On the other hand, the recommended
dehiscence, we found a statistically signifi- technique of fascia closure involves the use of
cantly greater number of people admitted in a continuous suture, with the suture length to
an emergency setting compared with people wound length ratio exceeding 4:1 (23, 24). In
admitted in an elective setting, which is in our study, we found no statistical relationship
Unauthenticated
Download Date | 5/12/19 10:07 AM
572 J. Kenig et al.
between dehiscence and incision techniques nases), which cause degradation of collagen,
(electric knife vs. scalpel) as well as fascia prolongation of the inflammation stage and
suturing techniques (single sutures vs. a con- activation of fibroblasts. Most published stud-
tinuous suture). Analysis of the literature re- ies concerning this subject support similar
veals that opinions are still mixed, although conclusions (14, 16, 21, 25, 27).
most papers, like ours, do not consider the
surgical technique (of course if properly per-
formed) to be a risk factor for wound dehis- Conclusions
cence (11, 25, 26). It is also interesting that
additional anti-eventration sutures were The results show that wound dehiscence is
placed during the first procedure in 25% pa- a complex process, influenced by both general
tients with wound dehiscence, but their role, and local as well as pre-, intra- and postopera-
in the light of EBM, is still to be determined. tive factors. Only concurrence of several factors
When it comes to opening the abdominal cav- can lead to this complication. Most risk factors
ity by means of a vertical incision, logistic re- do not depend upon the surgeon but mostly on
gression analysis did not prove this technique the patient’s gender, age, type of disease
to be a risk factor, but this results solely from treated in an emergency setting, and steroid
the study methodology. Similarly as in other use. The most important risk factor for wound
papers, 86% of patients in the group with dehiscence is surgical site infection. Therefore
wound dehiscence underwent laparotomy in the surgeon, along with the entire team caring
the midline of the body (11, 24). for the patient, should make every effort to
The most important risk factor for wound reduce the risk of this complication.
dehiscence remains surgical site infection. Ac- One should also remember that a knowledge
cording to published data, it is associated with of the above factors may allow a more detailed
an inflow of bacterial metalloproteinases and preoperative assessment and more informed
endotoxins (stimulating production of collage- patient’s consent to the surgery.
references
1. Jones V, Bale, S, Harding, K: Acute and chronic 10. Richards PC, Balch CM, Aldrete JS: Abdomi-
wounds. Wound care essentials: Practice principles. nal wound closure. A randomized prostpective
Philadelphia: Lippincott, Williams, & Wilkins study of 571 patients comparing continuous vs.
2004. interrupted suture techniques. Ann Surg 1970;
2. Szmidt J, Kużdżał J (red.): Podstawy chirurgii, 238: 1983-89.
tom 1. Wydawnictwo Medycyna Praktyczna 2010. 11. Van Ramshorst GH, Nieuwenhuizen J, Hop
3. Carlson MA: Acute wound failure. Surg Clin WCJ: Abdominal Wound Dehiscence in Adults:
North Am 1997; 77: 607-36. Development and Validation of a Risk Model. World
4. Burger JW, Van’t Riet M et al.: Abdominal inci- J Surg 2010; 34: 20-27.
sions: techniques and postoperative complications. 12. Seleh M, Hill A, AG Hill: Smoking is a major
Surg 2002; 91: 315-21. risk factor of wound dehiscence after midline ab-
5. Yahchouchy-Chouillard E, Aura T et al.: Incisio- dominal incision; case-control study. ANZ J Surg
nal hernia related risk factors. Surg 2003; 20: 2009; 79: 247-50.
3-9. 13. Penninckx FM, Poelmans SV, Kerremans RP et
6. Van’t Riet M, Steyerberg EW, Nellensteyn J et al.: Abdominal wound dehiscence in gastroentero-
al.: Meta-analysis of techniques for closure of mi- logical surgery. Ann Surg 1979; 189: 345-52.
dline abdominal incisions. Br J Surg 2002; 89: 14. Pavlidis TE, Galatianos IN, Papaziogas BT et
1350-56. al.: Complete dehiscence of the abdominal wound
7. Ceydeli A, Rucinski J, Wise L: Finding the best and incriminating factors. Eur J Surg 2001; 167:
abdominal closure: an evidence-based review of the 351-54.
literature. Curr Surg 2005; 62: 220-25. 15. Humar A, Ramcharan T et al.: Are wound
8. Afzal S, Bahir mm: Determinants of Wound complications after a kidney transplant more com-
Dehiscence in Abdominal Surgery in Public Sector mon with modern immunosuppression? Transplan-
Hospital. Annals 2008; 14(3): 110-14. tation 2001; 72: 1920-23.
9. Hodgson NC, Malthaner RA, Ostbye T: The search 16. Keill RH, Keitzer WF, Nichols WK et al.: Abdo-
for an ideal method of abdominal fascial closure: A me- minal wound dehiscence. Arch Surg 1973; 106:
ta-analysis. Ann Surg 2000; 231: 436-45. 573-77.
Unauthenticated
Download Date | 5/12/19 10:07 AM
Risk factors for wound dehiscence after laparotomy – clinical control trial 573
17. Niggebrugge AH, Hansen BE, Trimbos JB et 22. Papaziogas BT et al.: Complete dehiscence of
al.: Mechanical factors influencing the incidence of the abdominal wound and incriminating factors.
burst abdomen. Eur J Surg 1995; 161: 655-61. Eur J Surg 2001; 167: 351-54.
18. Kihara A, Kasamaki S, Kamano T et al.: Abdo- 23. Sowula A, Wermiński K, Groele H: Wytrzewie-
minal wound dehiscence In patients receiving long- nie a technika chirurgiczna. Pol Przegl Chir 2001;
term steroid treatment. J Int Med Res 2006; 34(2): 73(5): 417-29.
223-30. 24. Gembal P, Grzegorczyk W, Grabowski B i wsp.
19. Col C, Soran A, Col M: Can postoperative ab- Wiad Lek 2006; 59 (11-12): 767-71.
dominal wound dehiscence be predicted? Tokai J 25. Gislason H, Gronbech JE, Soreide O: Burst
Exp Clin Med 1998; 23: 123-27. abdomen and incisional hernia after major gastro-
20. Riou JP, Cohen JR, Johnson H Jr.: Factors intestinal operations-comparison of three closure
influencing wound dehiscence. Am J Surg 1992; techniques. Eur J Surg 1995; 161: 349-54.
163: 324-30. 26. Greenburg AG, Salic RP, Peskin GW: Wound
21. Webster C, Neumayer L, Smout RE et al.: Na- dehiscence. Pathophysiology and prevention. Arch
tional Veterans Affairs Surgical Quality Improve- Surg 1979; 114: 143-46.
ment Program. Prognostic models of abdominal 27. Armstrong CP, Dixon JM, Duffy SW et al.:
wound dehiscence after laparotomy. J Surg Res Wound healing in obstructive jaundice. Br J Surg
2003; 109: 130-37. 1984; 71: 267-70.
Received: 29.10.2012 r.
Adress correspondence: 31-202 Kraków, ul. Prądnicka 35/37
Unauthenticated
Download Date | 5/12/19 10:07 AM