Wound Dehiscence

Pathophysiology and Prevention

A. Gerson Greenburg, MD, PhD; Richard P. Saik, MD; Gerald W. Peskin, MD

\s=b\ A review of 32 abdominal wound dehiscences in a five-year primary surgical procedure are given in Table 1. The small bowel
period shows an incidence of 0.51%. Important factors are or colon involved in 17 cases (53%). In reference to the original
was
preexisting pulmonary disease, "malnutrition," intraoperative operation, dehiscence occurred in eight "clean" cases (25%), 19
contamination (often minimal), gastrointestinal distention, and "clean-contaminated" cases (59%), and five "dirty" cases (16%). A
aggressive tracheobronchial toilet in the postoperative period. clean-contaminated case is one in which the gastrointestinal (GI)
Incision direction and type of closure have little influence on tract is opened in a planned or unplanned fashion in what
dehiscence rates. Wound dehiscence results in a substantial otherwise is a clean case. For purposes of this study, electively
prolongation of hospital stay. Promptly recognized and treated, prepared colon resections are in this category.
wound dehiscence is no longer a highly lethal complication. Neoplasia was the primary pathology in six patients (19%),
(Arch Surg 114:143-146, 1979) primary inflammatory disease in seven (22%), trauma in two (6%),
acute GI bleeding in seven (22%), and some component of GI
obstruction was present in seven (22%). The associated diseases
As modern surgical techniques have evolved and surgical were as follows: severe obstructive pulmonary disease, ten; obesi¬
-¿ .science has developed, the basis for wound dehiscence ty, eight; hepatic insufficiency (cirrhosis), seven; malignancy, six;
has variously been attributed to systemic or local factors.1*5 cardiac, six; major weight loss (greater than 20%), five; and
This study was undertaken to define factors associated diabetes, two. Two of the patients were operated on for morbid
with dehiscence in terms of systemic, local, and patient obesity and two were receiving chronic steroid medications for
their pulmonary problems.
management influences that might identify the patient at Midline incisions were used in 21 patients (66%) and transverse
high risk and thereby provide the basis for a rational in 11 (34%). This corresponds in distribution to incisions in general
management approach. use on our services. Four patients in each group had retention
sutures (25%), corresponding to the general incidence of use.
DATA AND RESULTS
Primary fasciai closure was accomplished in all wounds using
A restrospective review of 32 abdominal wound dehiscences nonabsorbable suture as indicated in Table 2. A running fasciai
occurring at our two major teaching hospitals between January closure was used in 18 (56%) and an interrupted closure in 14 (44%).
1973 and January 1978 was accomplished. Only dehiscences occur¬ Table 2 gives the closures and represents essentially the distribu¬
ring on the general surgical service were included. For our tion of current practice in our institution.
purposes, wound dehiscence is defined as a wound disruption with No differences could be detected between elective and emergen¬
evisceration occurring in the immediate postoperative period cy groups based on day of week or time of day. Mean duration of
requiring closure to reduce the extraperitoneal bowel. During this surgery was identical for both groups, 2.9 hours. Intraoperative
time, 6,250 intra-abdominal procedures were performed yielding antibiotics by irrigation were used equally in elective and emer¬
an incidence of dehiscence of 0.51%. There were 27 men and five gency cases independent of classification. Postoperative antibiot¬
women in the series, with an average age of 60.68 ± 14.53 years ics were used three times more frequently in emergency cases, yet
(range, 23 to 88 years). Indications for exploration were emergent frank infection was present in only 16% at original procedure.
in 15 cases (47%) and elective in 17 (53%). The organs involved at Recall, however, that 59% of the patients had their GI tract
opened, with only nine of 19 (47%) receiving antibiotics.
Accepted for publication July 12, 1978. At the time of dehiscence, 22 of the patients (68%) had positive
From the Departments of Surgery, Veterans Administration Hospital
and the University of California, San Diego. cultures and ten of these were described as grossly purulent-45%
Reprint requests to Department of Surgery, Veterans Administration of positive and 31% of the total population. Clean-contaminated
Hospital, 3350 La Jolla Village Dr, San Diego, CA 92161 (Dr Greenburg). cases demonstrated positive cultures in a ratio of 3:1 over the

Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/04/2015

 grossly contaminated cases. Recovered organisms were as epidural anesthesia have been used for this high-risk group,
follows: especially for elective procedures.
Escherichia coli 13 Prolonged hospitalization is the hallmark of the morbidity
Barteroides fragilis 12 associated with dehiscence. Ten patients (31%) required less than
29 days of total hospital stay, whereas 22 patients (69%) were
Staphylococcus aureus 5
Klebsiella 3 hospitalized for 30 days or more, mean stay, 45 days with a range
of ten to 90 days. There is one death attributed directly to the
Streptococcus faecalis 3
dehiscence complication occurring in a patient undergoing intesti¬
Salmonella typhi 1
7 nal bypass for morbid obesity (3.1%). Two other patients died of
"Mixed," more than three types
respiratory failure not at all related to the dehiscence or infection.
Dehiscence occurred between two and 11 days, mean, 7.03 days. Six Of the remaining 29 patients, 72% were known to be alive at the
dehiscences occurred at four or less days (19%), five at five or six time of review and eight have been lost to follow-up. Ventral
days (16%), and 21 at seven or more days (66%). hernias have been noted in three of the 21 patients available for
All dehiscences were reclosed in the operating room. Description evaluation and all have been repaired primarily. Interestingly, all
of the findings at that time showed intact sutures with fasciai three were in wounds closed by fasciai reapproximation and not
disruption in 22 patients (68%), necrotic infected fascia in six the through-and-through closure. In the latter group, no hernias
(19%), and a broken running suture in four (13%). The latter, three are yet apparent.
of four, occurred in less than four days. Dehiscence repair was with
through-and-through sutures in 25 patients (78%) and fasciai COMMENT
resuturing in seven (22%). The closures with through-and-through Abdominal wound dehiscence continues to haunt the
sutures were different in 70% from the original closure: 18 of 25
(72%) were wire and seven of 25 (28%) were nylon. Four wounds
general surgeon. With newer and more sophisticated pre¬
were reclosed with interrupted fasciai sutures and three with
operative evaluation and preparation techniques the inci¬
dence seems to have stabilized at about 0.5%. However,
running fasciai sutures, 57% exactly as the primary closure and
43% different. In none of these latter seven closures was gross since the report of Halasz1 in 1968, there seems to be no
purulence or necrotic fascia observed. substantial advance despite the availability of newer
Potentially important postoperative events are given in Table 3. suture material and a wide ranging retinue of closure
In the patients with GI distention, two had no nasogastric tube techniques.
and three had the tube discontinued at less than 36 hours Systemic and local factors seem to exert the same
postOperation. One patient had delirium tremens in the immediate influences now as they did ten years ago. The considerable
postoperative period. No patient's dehiscence was preceded by a decrease in mortality associated with dehiscence is proba¬
major wound infection. In those dehiscences associated with bly more a reflection of better perioperative and postoper¬
grossly purulent drainage, an intra-abdominal abscess was the ative care rather than technical advances in wound closure.
cause. Leakage of this serosanguinous ascites was noted in only six
The current study further indicates that it is exactly these
patients prior to presentation with a wound disruption.
Of importance in this series is the fact that all patients had their factors that may also contribute to the dehiscence rate.
operation under general endotracheal anesthesia despite the fact The need for and the implementation of assertive
that about one third of them had substantial obstructive pulmo¬ tracheobronchial toilet, including nasotracheal suction and
nary disease. Equal numbers of elective and emergency proce¬ intermittent positive pressure breathing with or without
dures were performed on these patients. In general, spinal or humidified air in the immediate postoperative period,
could be in part responsible for the wound's disruption.
More than one half of our patients were in this category.
Table 1.—Organs Involved at Primary Operation Increased intra-abdominal pressure, forced coughing, and
strain on abdominal wounds independent of direction
Site_Total_Elective_Emergency
Colon 11 secondary to these measures and accompanied by early
Small bowel ambulation can be viewed as factors contributing to dehis¬
Appendix cence. That is not to say these should be avoided. Rather,
Gastroduodenal the perioperative management of the patient at risk must
Biliary be perfect in every way so that should the problem of
Aorta dehiscence appear, its effects are minimized.
Portal vein At obvious risk is the older man undergoing a colonie
Spleen
procedure or an emergency abdominal operation that is

Table 2.—Original Fasciai Closure Table 3.—Postoperative Events

Potentially Contributing to Dehiscence
Type Total Midline Transverse
Running Events _No. (%)
Polypropylene 17 12 On mechanical ventilator > 24 hr 13 (41)
Nylon Aggressive tracheobronchial toilet,
Interrupted nasotracheal suction,
Wire intermittent positive pressure breathing, etc 18 (56)
Polyester Pneumonia and atelectasis 10(31)
Polypropylene Ileus, gastrointestinal distention
_ _
7 (22)

Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/04/2015

 complicated by or involves entrance into the GI tract. The days address the need for more attention to the technical
so-called "clean-contaminated" case must be recognized aspects of wound closure. Fasciai disruption with "intact
and appropriate intraoperative irrigations used. This is sutures" found in the absence of fasciai necrosis may also
comparable to the information in the prospective report by be a technical problem related to poor suture placement.
Goligher et alfi of 1975. There is little firm evidence to One could argue that broken sutures or "pulled through"
support antibiotic irrigation as superior to balanced salt running sutures would not occur if an interrupted closure
solution. With the high incidence of infection in the technique was adopted. That we have seen as many
clean-contaminated cases, 15 of 19 cases (79%), it seems disruptions with either technique tends to discount the
reasonable to use antibiotic irrigation for this situation. argument. Dehiscence is probably a function of more than
The patient with severe chronic obstructive pulmonary closure technique!
disease obviously constitutes an increased risk and must be Failure to adequately decompress or maintain GI decom¬
managed more carefully. These patients often require pression is yet another aspect of management to be
surgical intervention despite their high-risk status. Vital considered. Gastric and small-bowel distention must be
to the success of the procedure is a well-organized, preoper¬ decompressed if present to provide safe abdominal closure.
ative, perioperative, and postoperative management plan Prevention of recurrent distention is crucial if one is to
designed to assure minimum compromise to respiratory prevent dehiscence. Enterotomy is not always necessary
function. Recall that this group often uses the abdominal for bowel decompression. A simple enterotomy can convert
muscles as accessory respiratory muscles and despite the a clean case to clean-contaminated and raise the risk of
advantages of transverse incision on pulmonary function, dehiscence. Passage of a long tube from above or "milking"
midline incisions should be considered. contents retrograde to stomach or forward to the colon
Preoperative tracheobronchial toilet is mandatory for often is feasible and effective decompression while obviat¬
this group of patients, particularly so in the elective ing an enterotomy. Maintenance of a nasogastric tube
situation. Obviously, this is not always feasible in the postoperatively prevents swallowed air and gaseous disten¬
emergency situation. If at all possible, regional or conduc¬ tion and is desirable.
tion block anesthesia should be used for elective and The technique of abdominal incision, especially of the
emergency procedures. Many midabdominal and lower fascia, is worthy of comment. In six of the eight most
abdominal procedures can be performed under adequate recent dehiscences, the electrocautery in coagulation mode
spinal or epidural anesthesia. In fact, high thoracic contin¬ was used in opening the fascia. In these six cases, there
uous epidural anesthesia, if available, can be used for most was obvious fasciai necrosis without purulence where the
abdominal surgery with the catheter left in place for use in suture material had pulled through at reexploration occur¬
postoperative pain management. ring on day 6 or 7. Of interest is that all of these were
The patient with obstructive pulmonary disease should emergency procedures in younger patients than the series
avoid general anesthesia if at all possible. Allowing such as a whole and all had retentions placed. Whether the
patients to breathe on their own and control pain provide a dehiscence was due to fasciai necrosis secondary to cauter¬
mechanism for prevention of the pulmonary complications ization or poorly placed sutures cannot be explicitly
that are associated with a large percentage of dehiscences. stated.
We recognize this is not a clear cut management situation. There seems to be no difference in the rate of dehiscence
There are cases where the patient with severe pulmonary relative to incisional direction. Dehiscence occurs in
disease also has cardiac compromise. In this subset of approximately the same ratio as these incisions are used.
patients, a general anesthetic with the attendant risks to Of interest is the total lack of paramedian incisions in the
pulmonary function is preferable because the cardiovascu¬ series.
lar system is under better control than it might be with a A large number of cases were either clean-contaminated
regional anesthetic. We recognize the difference of opinion or dirty, 75%, and the incidence of infection at time of
and propose this management as a mode, not necessarily dehiscence was 68%. Obviously, infection plays some role in
the mode. the pathogenesis of the dehiscence. The use of routine
Malnutrition either as obesity or weight loss is yet irrigation with or without antibiotic irrigation in these
another factor to be dealt with in the area of prevention. If cases is advocated. Our preference is to use antibiotic
one considers obesity as a form of malnutrition, then 13 irrigation. In the patient with pulmonary compromise,
patients (41%) of this series can be so classified. Converting recent infection, substantial infection, or major blood loss,
the wasted anorectic patient to positive nitrogen balance special attention to wound closure details is in order. The
preoperatively in the elective situation can only help older patient with these factors present is at particular risk
wound healing. The obese patient is a recognized risk for for dehiscence, especially if pulmonary complications occur
wound infection and very little can be done to close the and vigorous tracheobronchial toliet is required. If at all
huge dead space between skin and fascia. The use of possible the avoidance of general anesthesia is desirable
suction drains in this area seems to afford some degree of for it tends to exaggerate pulmonary difficulties.
protection in this situation. Not surprising is that our results are quite similar to
That four of our dehiscences were associated with "bro¬ those of Keills in that a single causative factor is difficult
ken suture" material, three of which were polypropylene, to identify. The pathophysiology of wound dehiscence is no
and that six dehiscences occurred between two and four doubt a combination of factors that, when taken together,

Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/04/2015

 produce the complication, whereas any one alone is not pulmonary disease and some degree of malnutrition. Open¬
quite sufficient to accomplish the deed. ing of the GI tract in an otherwise clean case seems to be
In recent reviews, two factors seem to be of major an additional factor. The mode of closure and the direction
importance in wound dehiscence. These are infections as of incision seem to bear little influence on the problem.
noted by Halasz,' Higgins et al,4 and Reitamo and Möller,7 Importantly, pulmonary complications and early removal
and postoperative pulmonary complications as emphasized of GI decompression tubes correlate well with the occur¬
by various authors.3'7's Our data would support these rence of dehiscence. Prolonged hospitalization is the main
concepts. morbidity, and mortality is low (3%), probably because of
We have no evidence to support a preference for midline improved management techniques. To the list of causes for
over transverse incisions nor can we implicate anemia or a dehiscence we can add those related to management. It is
particular underlying cause. What we do recognize is the an enigma that techniques used for improving patient
patient at risk and the profile of this patient is as follows: survival also contribute to morbidity.
man, older than 60 years of age, malnourished, with
moderate to severe obstructive pulmonary disease who is This research was supported in part by a grant from the Veterans
Administration Medical Research Service.
subjected to a colon operation or in whom the GI tract is
entered at emergency procedure. In this patient, develop¬ References
ment of considerable postoperative pulmonary complica¬ 1. Marsh RL: Factors involving wound dehiscence: Study of 1,000 cases.
tions, pneumonia and atelectasis, requiring vigorous JAMA 155:1197-1200, 1954.
tracheal-bronchial toilet is most frequently associated with 2. Miles RM: The etiology and prevention of abdominal wound disruption:
An analysis of 177 cases. Am Surg 30:566-573, 1964.
wound disruption. 3. Halasz NA: Dehiscence of laparotomy wounds. Am J Surg 116:210-214,
Modern surgical management has made it possible to 1968.
treat effectively and efficiently a complication previously 4. Higgins GA Jr, Antkowiak JG, Esterkyn SH: A clinical and laboratory
associated with high mortality. Wound dehiscence, when study of abdominal wound closure and dehiscence. Arch Surg 98:421-427,
1969.
properly managed, is no longer a lethal situation. 5. Keill RH: Abdominal wound dehiscence. Arch Surg 106:573-577, 1973.
6. Goligher JC, Irvin TT, Johnston D, et al: A controlled clinical trial of
CONCLUSION three methods of closure of laparotomy wounds. Br J Surg 62:823-892,
1975.
Wound dehiscence most frequently in older male
occurs 7. Reitamo J, M\l=o"\ller C: Abdominal wound dehiscence. Acta Chir Scand
abdominal 138:170-175, 1972.
patients undergoing surgery for colon or small- 8. Alexander HC, Prudden JF: The causes of abdominal wound disrup-
bowel problems who have associated chronic obstructive tion. Surg Gynecol Obstet 124:1223-1229, 1966.

Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/04/2015