Medical Science Publication No. 4, Volume 1
WOUND HEALING IN PATIENTS WITH SEVERE
BATTLE WOUNDS AND RENAL DYSFUNCTION*
STANLEY M. LEVENSON,M. D.
I am going to restrict my remarks this morning to a discussion of someof the problems of wound healing encountered among a group of patientssimilar to those just described by Major Meroney-that is, patients withextensive wounds and serious renal dysfunction. When I arrived in Japanand Korea in January, 1953, it was generally considered by most of theChiefs of Surgery and Surgical Consultants that delays in wound healingwere infrequent among patients with battle injuries except among thosewho also had acute renal failure. The physicians at the Renal Center werealso of the firm opinion that wound healing was impaired in these latterpatients and that wound complications contributed significantly to theirhigh mortality (50-60 percent). To obtain some specific data in this regard,all the clinical and autopsy records of the 70 patients admitted to theRenal Center from its opening in the spring of 1952 through the middleof February 1953 were reviewed in collaboration with Captain Paul Teschan.
The records of 21 of the patients were inadequate for analysis. Sevenof the remaining 49 records were discarded because of very short survivaltimes of the patients. Forty-two records, then, were deemed adequate foranalysis. However, it should be mentioned that the progress notes werewritten by internists, that the records were not specifically directedtowards problems of wound healing, that bacteriologic studies were inadequateand that only casual attention was paid to the wounds at autopsy.
Among the 42 patients whose records were analyzed, gross impairmentof wound healing was noted in 31. The term "impairment" is usedin a broad sense and is not meant to imply a specific or nonspecific defectin wound healing. Mortality among the patients with impaired wound healingwas high; wound complications were among the more frequent and more importantcauses of death (table 1).
Two general types of wounds were present in these patients-wounds closedprimarily, such as laparotomy incisions, and wounds left open after débridement.The time of secondary closure of the
*Presented 21 April1954, to the Course on Recent Advances in Medicine and Surgery, Army MedicalService Graduate School, Walter Reed Army Medical Center, Washington, D.C.
298
Table 1. Wound Healing in Patients with RenalDysfunction
Group | Number of patients | Number of deaths | Average day of death |
Patients with unimpaired wound healing | 11 | 0 | ----- |
Patients with impaired wound healing | 31 | 16 | 15 |
latter wounds was delayed in patients with renal failure. In patientswithout renal dysfunction, such wounds were usually closed 5 to 10 daysafter injury. Patients with renal dysfunction were often at their sickestduring this time and secondary closure was rarely carried out during thisperiod. Consequently, the wounds of patients with renal dysfunction remainedopen for significantly longer periods of time than in patients withoutrenal dysfunction. Similarly it should be pointed out that these patientswere among the most seriously injured and usually were in severe shockin the early period after injury. Consequently, it is possible that underthese circumstances débridement might have been inadequate in certainof these patients. There is also some question as to the quality of thelater surgical care of these patients.
Among the 31 patients with gross impairment of wound healing, the openwounds were often described as indolent, with granulation tissue absent,or when present, soggy and edematous. In a number of these patients, progressivenecrosis and suppuration of the wounds was described. I will not take timeto show any pictures of such wounds, since you have already seen many illustrationsof wounds just before and just after débridement and I am sure thatthe speakers this afternoon will present pictures of wounds in the laterstages.
Among the 42 patients whose records were analyzed, there were 28 whohad laparotomies (table 2). Six of these were in the group withunimpaired wound healing, and, as indicated by group classification, noneof these wounds ruptured. Among the 31 patients with impaired wound healing,22 had laparotomies. There were five abdominal wound dehiscences inthis group. This is a high incidence of wound rupture and is apparentlyhigher than occurred in patients with serious battle wounds but withoutrenal dysfunction. However, this cannot
Table 2. Wound Healing in Patients with RenalDysfunction
Group | Patients | Laparotomies | Dehiscences |
Patients with unimpaired wound healing | 11 | 6 | 0 |
Patients with impaired wound healing | 31 | 22 | 5 |
299
be stated with certainty at this time, since no overall systematic tabulationof wound healing among the battle casualties of the Korean conflict hasbeen made. In this regard, there has been considerable difference of opinionamong the surgical chiefs and consultants as to the incidence of postoperativehernias in laparotomized patients; opinions have varied from "veryfew" to "very many."
The actual incidence of impaired healing of laparotomy wounds amongthe patients with renal dysfunction may well be higher than indicated bythe figures of wound ruptures. Once it appeared that these patients mightbe having difficulty in wound healing, sutures were left in for many weeks.Under this circumstance, the wounds of some patients were described asheld together only by the retention sutures, with no apparent healing havingoccurred. Similarly, a number of patients whose laparotomy wounds lookedgood and appeared to be healing normally came to autopsy 7 or more daysafter injury. No special examination of the abdominal incision was maderoutinely, but in an occasional instance when the sutures were removedor an attempt was made to biopsy the wound, the wound fell apart.
Among the five patients with actual wound rupture, a possible localreason for the dehiscence was apparent in three (table 3). In one, dehiscenceoccurred on the ninth postoperative day and examination revealed an extensivehematoma in the wound; in another, dehiscence occurred on the tenth postoperativeday-peritonitis and an infected abdominal wound were present; in the third,dehiscence occurred on the fourteenth day-severe peritonitis with asciteswas present. No specific local reason for dehiscence was noted in the othertwo patients. In one of these two, dehiscence occurred on the eleventhday; in the other, on the sixteenth day. Disruption this late postoperativelyis unusual and would suggest a definite delay in the healing process. Atthe time of the dehiscence, this last patient was emaciated. The abdominalwound was resutured and the patient lived for 9 more days. At autopsy,the resutured wound was described as follows: "There is a longitudinalmid-line abdominal surgical incision measuring approximately 12 cm. inlength around which radiate multiple
Table 3. Wound Healing in Patients with RenalDysfuntion, Dehiscences of Laparotomy Wounds
Number of patients with laparotomies | 28. |
Number of dehiscences | 5. |
Apparent "local cause" for dehiscence | 3. |
Peritonitis | 1; 14th day. |
Hematoma | 1; 10th day. |
Wound infection | 1; 9th day. |
No apparent "local cause" for dehiscence | 2; 11th and 16th days. |
ascites300
superficial draining sinuses measuring 2 cm. in length. Granulationtissue lines the upper half of the surgical incision which presents anopening measuring 5 x 1 cm. The base of this sinus tract is lined by denseyellow-red granulation tissue. Wire sutures support the incision."Two more of the patients whose ruptured laparotomy wounds were resutureddied, but the wounds were not specifically examined at autopsy.
Two of the five patients with dehiscences lived (table 4). The resuturedwound apparently healed uneventfully in the patient whose dehiscence hadpresumably been secondary to the hematoma. The other patient who survivedwas the patient whose wound had ruptured 11 days postoperatively with nolocal cause for dehiscence being apparent. After resuturing, he developedperitonitis and intestinal obstruction. These complications were consideredby those caring for the patient as subsequent to the dehiscence, ratherthan present before, and perhaps etiologically important for, the firstdehiscence. A third laparotomy was performed on this patient through anotherincision 19 days after the resuturing of the first laparotomy wound. Localizedintraperitoneal abscesses and obstructing adhesions were found. One weeklater, purulent material drained from the first and second abdominal incisions.Shortly thereafter, a fecal fistula appeared in the other abdominal incision.The patient had lost a considerable amount of weight (over 30 pounds) andwas transferred to a general hospital in Japan for further therapy.
Table 4. Wound Healing in Patients with RenalDysfunction.
Number of secondary sutures of dehisced laparotomy wounds | 5 |
Number of patients surviving | 2 |
"Normal" healing of resutured wound | 1 |
"Abnormal" healing of resutured wound | 1 |
Number of patients dying | 3 |
"Abnormal" healing of resutured wound | 1 |
Quality of wound healing unknown | 2 |
A comparison between the two groups of patients (those patients withunimpaired wound healing versus those with impaired wound healing) wasmade to determine some of the factors which might have had some bearingon the observed differences in wound healing. It appears that those patientswith impaired wound healing were in general the more seriously injured,the sicker, the more uremic, required more dialyses, had greater changesin water and electrolyte metabolism with associated edema and/or dehydration,lost more weight, had more severe infections, and showed a higher incidenceof abnormal bleeding. The severity of anemia could not be adequately
301
estimated. In the majority of patients multiple factors were operativeand presumably interrelated. Serious uremia, marked malnutrition and severeinfection were the most frequent factors.
Severity of the Renal Dysfunction: Dialysis
Renal dysfunction was greater in the group with delayed wound healing.As you can see in table 5, the average time of diuresis among these latterpatients was 12 days after injury as opposed to the average time of 5 daysafter injury of 10 of the 11 patients with unimpaired wound healing. Inaddition, 11 patients with impaired wound healing died 5 to 19 days afterinjury without ever diuresing. Similarly, whereas only 3 of the 11 patientswith unimpaired wound healing had maximum levels of plasma nonprotein nitrogenover 275 mg. per 100 cc., two-thirds of the group with impaired healinghad such high levels.
Table 5. Wound Healing in Patients with RenalDysfunction.
Group | Number of patients | Day of diuresis | Number of patients with maximum blood NPN over275 mg. Percent |
Patients with unimpaired wound healing | 10 | 5 | 2 |
Patients with impaired wound healing | 31 | *12 | 21 |
*Includes 11 patients who died on the 5th to 19th dayswithout ever diuresing.
Among the patients with apparently unimpaired wound healing, there wasonly one who required dialysis (table 6). This was a Korean soldier withwounds of the face and neck who developed a severe hemolytic reaction toa transfusion of 1,000 cc. of blood during his initial operative treatment.Following this, he developed oliguria and was transferred to the 11th EvacuationHospital on the third post-injury day. During a prolonged period of oliguria(19 days) and uremia, three hemodialyses were performed. The neck and face
Table 6. Wound healing in Patients with RenalDysfunction
Group | Number of patients | Number of patients dialyzed | Number of dialyses |
Patients with unimpaired wound healing | 10 | 0 | 0 |
Patients with impaired wound healing | 31 | 24 | 56 |
302
wounds showed apparently normal and entirely satisfactory healing. Thesewounds were in well vascularized areas in which healing is usually favorable.
Among the 31 patients with apparently impaired wound healing, therewere 24 who required dialyses. The average number of dialyses in thesepatients was about 2.3. At the moment, there is no specific evidence toimplicate the dialysis procedure per se as an important factor inthe pathogenesis of the impaired wound healing-e. g., there were a numberof patients with impaired wound healing who did not have dialyses. Thefact that many more patients in the impaired healing group were dialyzedthan in the group with unimpaired healing may be simply indicative of theseverity of the hyperkalemia and uremia of the former group. On the otherhand, one cannot definitely rule out the dialysis procedure itself as animportant factor-e. g., what is the washout of the water-soluble vitamins(specifically ascorbic acid) during dialysis?
Malnutrition. Weight loss is a constant feature of the patientswith serious injuries and renal dysfunction. The average weight loss wasgreater among those patients with impaired wound healing. Weight lossesof 20 to 25 pounds were common among the group with unimpaired healing,while among the group with impaired healing 30- to 40-pound weight losswas not uncommon (table 7). What the weight loss specifically representsin terms of body tissue, water, fat, etc., is not known. The first weightsrecorded are those on admission of the patient to the 11th Evacuation Hospital.At this time, many of the patients were presumably waterlogged. Insensiblewater loss must be an important factor in the weight loss of these patients,since prolonged hyperpnea is a common feature. However, examination ofmetabolic data available in a few patients reveals a large nitrogen "loss"(i. e., NPN accumulation in the body water, NPN lost by dialysis, and NPNexcreted in the urine). In one patient, this amounted to about 45 gm. Nper day, which represents the daily breakdown of about 2.5 pounds of bodytissue (excluding fat).
Table 7. Wound Healing in Patients with RenalDysfunction
Group | Number of patients | Weight loss (pounds) |
Patients with unimpaired wound healing | 11 | 20-30 |
Patients with impaired wound healing | 31 | 30-45 |
Infection. Infection was one of the major complications amongthe severely injured patients with renal dysfunction. I am not going to
303
spend much time on this aspect of the problem because it is discussedin three other papers. However, I want to indicate some of the data whichapply to this particular series of patients. Among the 31 patients withimpaired wound healing, wound infection was almost universally present.Infection, either in the wound or elsewhere, is listed among the causesof death in all but 1 of the 16 patients in this group who died. Peritonitisis listed four times; intra-abdominal abscesses, six times; spreading infectionof peripheral wounds, four times; severe bronchopneumonia, four times;septic infarcts (lung, kidney, etc.) twice, and empyema twice.
In view of certain observations of an increased migration of bacteriaacross the intestinal wall in uremic dogs, the autopsy records were examinedto see whether there have been any instances of peritonitis in the absenceof intra-abdominal injury. No instance of peritonitis in the absence ofa previous laparotomy was found. There were one or two instances of mildperitoneal infection in patients with a "negative" laparotomy,but the possibility of undiscovered intra-abdominal injury cannot be ruledout.
It was evident from reviewing the records of these patients that woundhealing was a very important complication among the patients with renaldysfunction. It would appear that in the majority of the patients multiplefactors were operative and presumably interrelated. However, it was apparentthat with the data at hand, no specific conclusion as to the relative importance,or interrelationship, of the various factors could be made. Studies todefine the problem specifically and thereby lead to improved prophylaxisand therapy were indicated. A start in certain of these will be presentedby other speakers later today.
Some of the problems which needed (and still need) solution are as follows:
1. What is the course of normal wound healing (open wounds, suturedwounds, etc.) in man?
There is a paucity of detailed correlated (clinical, histologic, bacteriologic,etc.) information regarding wound healing in man. There are very few controlledclinical studies and none specifically applicable to the problems at hand.Most of the specific data regarding wound healing have been obtained inanimals. The difference in wound healing among various species makes itimperative that caution be used in directly relating the results of animalexperiments to man.
2. What is the effect of the magnitude of injury on wound healing, and,if an effect is present, to what factors may it be attributed? What isthe clinical significance of the "catabolic" reaction to injury?
Very few objective data concerning the physiological and clinical sequelaedirectly attributable to the early "catabolic" reaction to in-
304
jury are available. There are conflicting opinions as to the harm resultingfrom this period, and depending on the viewpoint taken, attempts are orare not made to reverse the process. Much of the conflict is due to thelack of objective indices of the benefits, or lack of benefits,of mitigating the early metabolic disturbances.
The injured man must heal his wounds for successful recovery; systematicobservations of the healing of wounds, traumatic, operative and experimental,would provide objective evidence in one important area as to the significanceof the "catabolic" period. For example, the seriously injuredindividual acts biochemically like a scorbutic in the first days and wellafter injury; does he also act like a scorbutic in regard to the healingof his wounds? Further, the intensity of the urinary nitrogen loss followinginjury may be decreased by the injection of testosterone propionate. Sincethe anabolic effects of testosterone are different for different tissues,what does the decrease in urinary nitrogen excretion mean in terms of woundhealing?
It has been postulated by some that no attempt be made to reverse the"catabolic" reaction because it is a "defense mechanism"to supply metabolites to the injured area. There is no concrete evidenceto support this. There is no reason, at the moment, to assume that theinjured area is necessarily more proficient than other tissues in "utilizing"the circulating metabolites. We have recently studied the healing of experimentallaparotomy wounds in normal and severely burned rats. Observations of thegross appearances, tensile strengths and histologic features of the incisionswere made. The healing of laparotomy wounds in the burned rats was significantlydifferent from that in the unburned controls. Epithelization was not affected,but there was a definite delay in the formation of granulation tissue inthe incisional wounds of the burned animals with a lag in the appearanceand maturation of the fibroblasts and the ground substance. The eventualnumber and amount of these two elements, however, did not appear to beaffected, and abundant granulation formed in the wounds of the burned ratsin time. In some of the burned rats the wound area appeared somewhat moreedematous than that of the controls. The incidence of wound infection wasalso somewhat higher among the burned animals.
3. Is there a specific effect (direct or indirect) of renal dysfunctionon wound healing? Or are the delays due to associated abnormalties in nutrition,water balance, ability to resist infection, etc.?
Various degrees of renal dysfunction will be produced experimentallyin a number of different ways. Emphasis will be directed toward simulatingthe clinical problem of "lower nephron nephrosis." The courseof wound healing in animals with renal dysfunction, untreated and treatedin a variety of ways, including dialysis, will be studied.
305
Observations on local and systemic infection and various immune responseswill be made. These data will be correlated with various nutritional andmetabolic measurements.
4. What is the basis for the apparent high incidence of wound infectionin the patients with renal dysfunction?
It is well recognized that infection, when present, is a detrimentto wound healing. A careful study of wound infection is important, notonly in the early post-injury period, but throughout the healing period.Why is wound infection so frequent, and so serious, in the severely woundedpatient with renal dysfunction? Does the malnutrition predispose to woundinfection, or does the wound infection accelerate the development of malnutrition?What is the ability of the seriously injured man in regard to antibacterialdefense? Following simple starvation, lymphoid tissue is markedly depleted;chronically protein-depleted rats are unable to synthesize certain antibodiesas well as normally nourished animals. What is the ability of the seriouslywounded man who is on a totally inadequate diet to form antibodies? Whatis the efficiency of phagocytosis, etc., in such an individual? Further,most of these patients may be on various antibiotics, certain of which,when given orally, may lead to nutritional disturbances under certain circumstances.
5. What are the effects of plasma substitutes and/or anemiaon wound healing?
It would appear that in many instances a combination of whole bloodand dextran (or some other plasma substitute) may be satisfactory for earlyreplacement therapy of shock. Under this circumstance, a certain degreeof anemia will be present at the time the patient is evacuated furtherto the rear. Ordinarily, surgeons feel that anemia per se is detrimentalto wound healing and will be inclined to transfuse such patients priorto secondary closure, etc. Is this a necessary or wise procedure (consideringpossible shortage of blood, transfusion reactions, etc.)? Is there a directeffect of anemia on wound healing or is there, perhaps, an indirect effect?Is hemoglobin a high-priority protein in the severely injured patient duringthe catabolic period and, if so, will protein be diverted from the healingwound to form hemoglobin if anemia is present? No conclusive data on theinfluence of anemia on wound healing in man are available; the data inanimals are controversial.
What is the effect of plasma expanders per se on wound healing?Data in this regard are meager. Rhoads and his co-workers observed thatwhereas there was a delay in the healing of abdominal wounds in hypoproteinemicedematous dogs, there was no delay in hypoproteinemic dogs given acaciaintravenously in amounts sufficient to eliminate the edema, but which,at the same time, accentuated the de-
306
crease in plasma protein concentration. Thorsen has reported no delayin the healing of incisional wounds in rabbits given dextran. We have observedno gross abnormalities in the healing of burns in patients who have receivedlarge amounts of dextran, but no special studies of the wounds were made.
6. What are the optimal prophylactic and therapeutic nutritional (dietary,hormonal, etc.) regimens for the wounded patient with or without renaldysfunction?
If the period of metabolic derangement persists, progressive nutritionaldeterioration with its consequent well known ill effects occurs. What isthe optimal nutritional (dietary, hormonal, etc.) care of these individuals?I will discuss this in detail in another paper.
Proposed Studies
From the foregoing it is evident that a study of the individual andhis wounds directed toward a comprehensive correlation and evaluation ofsystemic and local phenomena is needed. Such a study will entail the useof a variety of technics, clinical, metabolic, bacteriologic and pathologic.The clinical studies should be supplemented by animal studies in a varietyof species.
Some Factors in Wound Healing Requiring Control
A. Systemic Factors.
1. Extent and sites of wounds.
2. Associated injuries (and/or illnesses).
3. Circulatory system (shock, sludging, etc.).
4. Metabolic and nutritional state (including anemia, antibiotics, allnutrients, etc.).
5. Plasma substitutes (primary and secondary effects).
6. Infection (including "resistance" of individual, etc.).
7. Blood clotting mechanism.
B. Type of Wounds.
1. Contaminated wounds, dérided and secondarily sutured (includingtime of secondary suture).
2. Contaminated wounds débrided and primarily closed.
3. Clean, incised wounds closed primarily (including type of closure, etc.).
C. Local Factors.
1. Extent and sites of wounds (including surrounding and supportingtissues; proximity to joints; direction of wound relative to lines of stress,etc.).
2. Blood supply (arterial and venous).
3. Infection (including antibiotics, etc.).
4. Wound edema or dehydration (local or systemic basis).
307
D. Medical and Surgical Care. The quality and types of medicaland surgical care are, of course, of paramount importance, but will notbe discussed in this paper.
In summary, this analysis of wound healing among patients with seriousbattle injuries and renal failure confirms the impressions of various physicansthat wound complications were frequent and important in these patients.Further, the analysis has indicated the complexity of the problem, somepossible interrelationships among various factors, and the need for concreteobjective study, clinical and experimental.
Discussion
COLONEL HANSON. Stimulatedby Dr. Levenson's visit to the Far East, a study of some of the problemsof wound healing was undertaken by the research team in Korea in collaborationwith the staff of the 406th Medical General Laboratory in Tokyo. Serialbiopsy materials obtained from about 19 patients were examined and an attemptwas made to evaluate certain histologic features such as the general stateof healing as a whole, growth of capillaries, proliferation of fibroblasts,production of reticulum and collagen.
Some of the preliminary conclusions may be mentioned. I think that wecan state that in the severely wounded casualty, the response to injurymay be lessened in patients with marked renal failure, in cases where thereis a diminished blood supply or local anemia, and in that group of caseswhere there has been a marked catabolic effect and severe weight loss.The wounds of such patients do not heal like those of the normal or lessseverely traumatized patients. However, we could not draw any conclusionsas to mechanisms involved, but it is likely that many mechanisms, probablyinterrelated, are involved.
MAJOR BALCH. One thing Dr.Levenson did not bring out quite clearly enough in his presentation wasthe character of the surgical care of the patient whose records he reviewed.I was not at the Renal Center at the time those cases were studied, butI understand from others who were there that at the time the Renal Centerwas first set up, it was not properly realized that a full-time surgeonwas needed in the care of these patients. Many of those patients were caredfor surgically by the general surgical staff of the hospital. Now, thosesurgeons had other duties and responsibilities and they were frequentlyquite busy and could not come for many hours or, perhaps longer, to takecare of the renal patients. So, evaluation of the complications that havebeen reported today, and I believe they are true complications, is complicatedvery much by the fact that it is likely that the surgical care was inadequatein those patients.
308
I would like to ask Colonel Hanson whether the factor of infection couldbe separated from delay in wound healing in the biopsy specimens?
COLONEL HANSON. In those caseswhere there was massive wound infection, the factor of infection couldbe separated from delays in wound healing, but in cases where there wasjust minor infection, it was questionable whether these factors could beseparated.
DR. HOWARD. I wonder if oneof the most important elements in the wound complications among these patientsis their primary surgical care? We have by definition selected a groupof patients who are critically injured, who are profoundly hypotensiveand in whom the primary surgical care is quite likely to be compromised.I wonder if inadequate débridement or hasty closure of an abdominalwound, due to our efforts to salvage life at the moment, might not be reflectingitself in later wound complications?
DR. LEVENSON. That is probablytrue, Dr. Howard, and I mentioned that briefly in my presentation. However,at present, most records I have seen are not adequate to enable objectiveevaluation of the adequacy of débridement. I think we all wouldassume that in many instances débridement was inadequate, and ina few instances foreign material (clothing, etc.) was found in secondarydébridements.
I am glad that Major Balch made his comment regarding the questionablequality of the later surgical care of these patients. I meant to mentionit, but inadvertently did not. In this regard I would like to make oneadditional comment. Cross-infection was one of the factors possibly importantin the high incidence of wound infection among this group of patients.These patients were kept on a single ward and it was not possible to carryout, either on the ward, in the dressing or operating rooms, procedureswhich have been advocated for the prevention of cross-infections.
