U.S. Army Medical Department, Office of Medical History
Skip Navigation, go to content

HISTORY OF THE OFFICE OF MEDICAL HISTORY

AMEDD BIOGRAPHIES

AMEDD CORPS HISTORY

BOOKS AND DOCUMENTS

HISTORICAL ART WORK & IMAGES

MEDICAL MEMOIRS

AMEDD MEDAL OF HONOR RECIPIENTS External Link, Opens in New Window

ORGANIZATIONAL HISTORIES

THE SURGEONS GENERAL

ANNUAL REPORTS OF THE SURGEON GENERAL

AMEDD UNIT PATCHES AND LINEAGE

THE AMEDD HISTORIAN NEWSLETTER

Abdominal and Thoraco-Abdominal Wounds

Medical Science Publication No. 4, Volume 1

ABDOMINAL AND THORACO-ABDOMINAL WOUNDS*

CAPTAIN ALVIN W. BRONWELL, MC
MAJOR CURTIS P. ARTZ, MC
AND
CAPTAIN YOSHIO SAKO, MC

Not until the middle of World War I were routine laparotomies performed for exploration of abdominal wounds produced by battle injuries. Wounds of the abdomen were treated by rest and sedation prior to that time. Information was obtained in World War II as to the preferred method of treatment of various types of abdominal injuries. During the final 6 months of the Korean conflict, essentially the same operative procedures were used in all of the forward hospitals.

Incidence

Before the armored vest was used in Korea, 19 percent of the wounds of the body were in the chest, and 11 percent in the abdomen; thus 30 percent of all wounds were in the trunk. There was a decrease of 10 percent in trunk wounds after the use of the armored vest; 8.7 percent of all wounds were in the thorax and 10.8 percent in the abdomen (1).

Diagnosis

Patients with intra-abdominal injuries usually arrive at the forward hospital with low blood pressure and a rapid pulse. The time interval front injury to admission, the extent of peritoneal contamination, and the amount of blood loss are factors in determining the degree of injury in casualties with comparable wounds. A rapid response to resuscitation is a good indication that the injury is not extensive. The severely wounded casualty responds slowly to resuscitative measures. Pain is of little diagnostic value because many of the patients have received intravenous morphine prior to admission. Pain is not the most prominent finding even in those patients who have not received morphine.


*Presented 22 April 1954, to the Course on Recent Advances in Medicine and Surgery, Army Medical Service Graduate School, Walter Reed Army Medical Center, Washington, D. C.


420

In taking the history, it is of value to determine the type of missile which produced the injury and the position of the patient when hit. After penetrating the skin, high-velocity missiles cause extensive destruction of internal organs. Low-velocity missiles of comparable size have a tearing action on the skin, with less internal destruction of tissue; or they may not penetrate the abdominal wall. Inspection of points of entrance and exit of the missile may suggest which organ or organs are damaged.

Abdominal distention is rarely present unless there is massive, intra-abdominal hemorrhage. Boardlike rigidity of the abdomen is seldom seen. When present, it is diagnostic of a perforated abdominal viscus. Patients with chest wounds or superficial abdominal wounds may have marked guarding or rigidity of the abdomen, even though there is no peritoneal irritation. Many patients with perforated abdominal viscus have a soft abdomen at the time of admission. During resuscitation, such patients usually develop rigidity of the abdomen. Generalized tenderness is usually an indication of intra-abdominal injury.

A silent abdomen on auscultation is a good indication of a perforated hollow viscus, resulting in leakage of intestinal contents. If peristalsis is present, along with other negative criteria, intra-abdominal injury is doubtful and the patient should be carefully observed.

Roentgenograms should be taken of all casualties suspected of having abdominal injury. They demonstrate the presence of intra-abdominal shell fragments, free air and retroperitoneal hemorrhage. One is not always able to localize shell fragments by roentgenograms in instances where fragments have entered through the back or lie close to the peritoneum, and where there is a wound of the perineum and buttocks. When a psoas shadow cannot be visualized on a roentgenogram, one should suspect the presence of retroperitoneal hemorrhage.

Peritoneal taps are of little value in establishing the diagnosis of intra-abdominal injury. An abdominal tap with positive findings is a definite aid; while a negative tap does not rule out injury to intra-abdominal organs.

Technics of Abdominal Operations

All patients in this series received pentothal induction and endotracheal gas-oxygen-ether anesthesia, and curare-like drugs. The skin of the abdomen is prepared by shaving, washing with a detergent containing hexochlorophene, and irrigating with saline solution.


421

Incisions

A long, muscle-splitting incision into the rectus muscle should be used for exploration of the abdomen. It should extend from the costal margin to the pubis. The abdominal incision should never pass through the site of injury; but it should be placed a considerable distance from it or, if necessary, on the opposite side. A systematic examination of the abdominal cavity should be made immediately after the peritoneum has been opened. When bleeding is present, the small intestine should be eviscerated to facilitate rapid exploration of the peritoneal cavity. The bleeding points can then be observed and the hemorrhage controlled immediately. When this has been completed, the contents of the peritoneal cavity should be thoroughly inspected from the cardia of the stomach to the peritoneal reflexion about the sigmoid. When abdominal bleeding is not a problem, a complete examination of the peritoneal cavity should be made with out eviscerating the small intestine.

Stomach

On many occasions, battle casualties have eaten a short time before injury. At operation they may still have a full stomach, even though vigorous attempts have been made to empty it. In such instances, it is advisable to incise the stomach and empty it. If this is not done, vomiting, aspiration and acute gastric dilatation may occur postoperatively. All wounds of the stomach require a thorough exploration of the posterior surface. This is accomplished by a transverse incision into the lesser sac through the gastrocolic omentum. The stomach can then be elevated in order that the entire posterior surface may be examined. Wounds of the stomach should be closed by two layers of interrupted silk sutures.

Duodenum

A perforating wound of the anterior surface of the duodenum always requires a thorough exploration of the posterior surface. Adequate exposure is necessary for an accurate repair of these wounds. The ascending colon and the hepatic flexure are mobilized. The duodenum can then be freed from its attachments and rotated medially, thus exposing the posterior surface of the second and third parts. Wounds in the second portion of the duodenum require careful dissection and repair to avoid injury to the common bile duct. It is well recognized that duodenal wounds heal poorly. All wounds should be sutured transversely to prevent a constriction of the lumen. A two-layer closure should always be made, and drains should be placed in the region of repair and brought out through a separate stab wound on the skin.


422

Liver

A small, superficial laceration of the liver in which there is no bleeding at the time of the operation requires only drainage. Small wounds which are bleeding should be sutured. Large, deep lacerations of the liver should be closed with a hemostatic gauze pack placed in the wound and edges approximated with wide, deep, figure-of-8 absorbable sutures.

Shell fragments are removed if they can be obtained without excessive trauma to the liver. Fragments which are embedded deep within the liver should not be removed.

All liver wounds should be drained, regardless of their size. Large Penrose tissue drains should be brought out through a stab wound in the midaxillary line and placed in each of the following areas. One should be placed in the foramen of Winslow, another anterior to the common bile duct, a third between the posterior surface of the liver and the diaphragm, and a fourth between the anterior surface of the liver and the diaphragm. Patients who are not drained frequently develop bile peritonitis.

Posterior wounds of the liver can be exposed by dividing the triangular and falciform ligaments. This permits the liver to be dropped and the posterior surface may be visualized. Perforation of the diaphram should be closed with interrupted sutures.

Gallbladder

Wounds of the gallbladder are treated by cholecystectomy. If the gallbladder has been injured, the common bile duct should be carefully examined. If the duct has been injured, it should be approximated over a T-tube. The T-tube should not be brought out through the anastomosis. All injuries of this area should be drained.

Spleen

All injuries of the spleen should be treated by splenectomy.

Small Intestine

The entire small bowel should be inspected from the ligament of Treitz to the cecum. At this time, bleeding points should be ligated. One can determine the type of treatment that is required upon completing the examination of the small intestine. In those instances in which there are only a few perforations of the bowel wall, a closure of the separate wounds should be made. Usually segments of the bowel with multiple perforations will require resection with an end-to-end anastomosis. Wounds of the small intestine should be closed transversely with a single layer of interrupted silk.


423

Cecum

One of the major problems in abdominal surgery is the treatment of wounds of the cecum. There is little or no agreement on the technic. Ileostomies and cecostomies are contraindicated because it is difficult to maintain electrolyte balance during evacuation. Primary closure of the cecum is frequently unsuccessful.

In small wounds limited to the cecum, tube cecostomies or suturing of the wall of the cecum to the skin are acceptable procedures. If a tube cecostomy is used, it is important to suture the cecum to the anterior peritoneum about the point of exit of the catheter. This will prevent peritoneal contamination, if there is leakage. In larger wounds of the cecum and ascending colon, it is advisable to perform a resection of the cecum with an end-to-end ileotransverse colostomy, or perform a resection of the cecum with end-to-side ileotransverse colostomy and single-barrel colostomy of the proximal colon.

Colon

A colostomy should be performed for all wounds of the colon. A single perforation of the colon should be treated by making a stab wound through the abdominal wall, bringing the injured area of the bowel out, and placing it over a glass rod as a loop colostomy. The wound of the colon is left open and later enlarged. Multiple wounds of the colon are treated by suturing the wound and making a defunctioning colostomy at the site of the proximal wound. In wounds of the descending colon, some surgeons prefer not to mobilize the descending colon but to suture both wounds and perform a proximal defunctioning colostomy in the transverse colon. Defunctioning colostomy should be brought out through two separate stab wounds which are at least 2 inches apart. Multiple perforations with massive destruction of the colon are treated with a resection of the involved segment. The divided ends of the colon are brought out through separate stab wounds at a convenient point.

Pancreas

Pancreatic injuries are rare. Hemorrhage and pancreatitis are the most frequent complications of wounds to the pancreas. All wounds of the pancreas should be drained with large rubber tissue drains placed about the injured area and brought out through a stab wound. The tail of the pancreas should be resected if extensively traumatized.

Retroperitoneal Injuries

Injuries to the retroperitoneal areas frequently cause large hematomas and extensive damage to muscle. The hematomas may be caused by perforation of a large blood vessel. If this occurs, the bleeding


424

should be controlled and the vessel should be repaired. Ligation of the vena cava, with associated retroperitoneal muscle damage, will frequently cause a secondary hemorrhage. Retroperitoneal muscle injury presents the problems of adequate excision of devitalized muscle and the control of hemorrhage from multiple bleeding points. It is difficult to expose the retroperitoneal muscle in order that an adequate débridement can be performed. When oozing occurs, it is necessary to pack the muscle bed to control the bleeding. This area should always be drained. Postoperative hypotension frequently occurs in patients who have retroperitoneal injuries, but will usually respond to adequate blood replacement. More blood is lost from the damaged muscle than is usually recognized.

Thoraco-abdominal Injuries

The maintenance of adequate oxygen exchange and the relieving of respiratory embarrassment make the chest the first priority in the treatment of thoraco-abdominal wounds. Hemothorax should be treated by thoracentesis and thoracotomies should be performed only on proper indications.

Sucking wounds of the chest should be débrided and closed. The abdomen should always be explored through a separate incision. All perforations of the diaphragm should be closed with interrupted silk sutures.

Abdominal Wall Closure

The peritoneum should be closed with running sutures of chromic catgut. The muscle, fascia, and skin should be approximated with through-and-through rubber-shod wire sutures. For better approximation, a few interrupted catgut sutures may be placed in the fascia between the wire sutures. More accurate skin approximation can be obtained with an occasional interrupted silk suture.

The technic of treatment of wounds of the rectum, kidney, ureter, bladder and arteries is a part of another discussion.

Results

During the final 3 months (1 May to 1 August 1953) of the Korean conflict, a statistical data sheet was made out on all general surgical casualties admitted to the 46th Surgical Hospital. The data collected from these records have been used to compile statistics for a detailed study of the casualty's care from the time of his admission to his discharge. During this period, 75 patients were admitted with abdominal wounds and 29 with thoraco-abdominal injuries (table 1). Nine


425

Table 1. Mortality and Mode of Evacuation

Type of wound

Total

Deaths

Mortality (percent)

Evacuation

Ambulance

Helicopter

Not stated

Abdominal

75

9

12. 0

40

27

8

Thoraco-abdominal

29

3

10. 3

19

8

2

Total

104

12

11. 5

59

35

10

deaths occurred among those with abdominal injuries, a mortality rate of 12 percent. Three deaths occurred in the patients with thoraco-abdominal injuries, a mortality rate of 10.6 percent. There is essentially no difference in the mortality rate of these two types of wounds.

Fifty-nine patients were evacuated by ambulance and 35 by helicopter. In 10 cases, the mode of evacuation was not stated. The percentage of casualties evacuated by helicopter in the abdominal group was slightly higher than in the thoraco-abdominal group. In most instances the helicopter was used to evacuate the most seriously wounded.

Time Interval

In the 75 abdominal casualties, the time from injury to admission to the hospital was 3.1 hours (table 2). The length of time required to prepare the patients for operation was practically the same. Thus there was an average time interval of 6.3 hours from injury until operation.

Table 2. Time Intervals

Type of wound

Number of cases

Hours injury to admission

Hours admission to surgery

Hours injury to surgery

Operative time (hours)

Abdominal

75

3. 1

3. 2

6. 3

2. 4

Thoraco-abdominal

29

4. 7

2. 7

7. 9

2. 3

The casualties with thoraco-abdominal injuries had a greater time interval from injury to admission to the hospital (4.7 hours), but a shorter period of time was required for resuscitation for operation (2.7 hours). Difference in the time interval was probably due to the greater number of them that were evacuated by ambulance. Essentially there was no difference between the two groups in the operative time (2.3 hours).


426

Average Amount of Blood Required in Resuscitation

An analysis of the amount of blood given to the abdominal and thoraco-abdominal patients during resuscitation was made (table 3).

Table 3. Average Blood Administered During Resusitation

Type of wound

Total number of cases

Blood prior to admission (cc.)

Blood prior to operation (cc.)

Blood during operation (cc.)

Average amount of blood first 24 hours (cc.)

Abdominal*

75

264

1,464

1,467

3,428

Thoraco-abdominal

29

103

1,308

1,187

2,867

*Patients receiving 40, 46, 52, and 56 pints excluded.

It was shown that the amount of blood (3 pints) given during the preoperative period to casualties sustaining abdominal wounds was approximately equal to the amount given during the operation. The patients who had thoraco-abdominal wounds received a slightly higher amount of blood preoperatively (1,308 cc.) than during the operation (1,187 cc.). About 7 pints of blood (3,428 cc.) was required in the first 24 hours by patients with abdominal wounds. Those with thoraco-abdominal wounds required 2,867 cc. of blood during this period. Patients who received 40, 46, 52, and 56 pints of blood respectively during their first 24-hour period were not included in these figures. It was felt that, by excluding them, a more accurate estimate of the average amount of blood required could be obtained. Most of the casualties received from 1 to 6 units of blood in the first 24 hours.

Causes of Death

There were nine deaths in the abdominal group (table 4). Five patients died from uncontrolled hemorrhage, two died of peritonitis and one each from aspiration pneumonia and pancreatitis. Two of the patients who died of uncontrolled hemorrhage died during operation. In one patient the common iliac artery was severed. He had a large retroperitoneal hematoma. Upon opening the peritoneum over the hematoma, active bleeding was controlled promptly; however, the patient went into immediate shock and died even though large quantities of blood were given, 19 pints in 4 hours. One patient died of cardiac arrest following prolonged hemorrhage; operation was delayed for 10 hours because of the heavy casualty load. The three other patients who died of uncontrolled hemorrhage had massive wounds of the muscle. They received 40, 52 and 56 pints of blood respectively in the first 24 hours. These patients had extensive muscle damage and illustrate the difficulty of controlling hemorrhage in such cases.


427

One patient died of aspiration pneumonia. During his evacuation by helicopter, he aspirated vomitus. Bronchoscopy was done on numerous occasions, but without improvement. His case points out the importance of emptying the stomach of its contents as soon as possible after injury. Two patients died of peritonitis. Both had injuries of the cecum. One had an additional injury to the duodenum, while the other had an additional injury of the small bowel. One patient died of pancreatitis; this patient might have been saved by a resection of the tail of the pancreas or better drainage of the area.

Thoraco-Abdominal Group

There were three deaths in the thoraco-abdominal group, one each from cardiac arrest, postoperative shock and hemorrhage from a lacerated liver (table 4). All died within the first 39 hours after injury. All had wounds of the right side of the diaphragm with associated liver damage. One patient had an extensive laceration of the liver which was repaired; and hemorrhage appeared to be controlled at operation.

Table 4. Causes of Death

Abdominal:

75 cases, 9 death

Uncontrolled hemorrhage
Peritonitis
Aspiration pneumonia
Pancreatitis

5
2
1
1

Thoraco-abdominal:

      29 cases, 3 deaths

Uncontrolled hemorrhage
Postoperative shock
Cardiac arrest

1
1
1

A secondary hemorrhage from the liver occurred several hours after operation and caused his death. Another patient had extensive injuries of the liver, duodenum and ileum. It was impossible to maintain a satisfactory blood pressure at any time during or following his operation. It was thought that continued hemorrhage caused his death. The remaining patient with extensive liver and chest injuries had a cardiac arrest during operation; the heart was massaged until return of normal rhythm. Twenty-nine hours after his operation, the patient had another cardiac arrest and died.

Negative Explorations of the Abdomen

No intra-abdominal injury was found in 36 (9.2 percent) of the 391 patients who had abdominal operations (table 5). Prior to operation, it is sometimes difficult to be certain that an abdominal viscus had not been perforated. This is especially true when the shell frag-


428

Table 5. Negative Explorations of the Abdomen

Investigators*

Number of laparotomies

Negative laparotomies

Percent negative laparotomies

Pearson, Tuhy, and Welch-1945, American-Northern Europe

290

23

7. 9

O’Gilvie-1942-43, British-Western Desert

247

42

17. 0

Edwards and Stead-1944, British-Italian Campaign

560

66

11. 8

Porritt-1944-45, British

4,319

740

17. 1

46th Surgical Hospital-1952-53, Korea

391

36

9. 2

*See references.

ments have entered the abdominal wall anteriorly and lie close to the peritoneum. Many times, fragments that have entered the back or the perineum present the same problem. When it cannot be established that a viscus has been perforated, an exploratory operation should be performed. Our experience compares favorably with that of Welch in World War II who reported negative laparotomies in 7.9 percent of 290 operations (2). In various campaigns of World War II, the British reported negative abdominal explorations varying from 11.8 to 17.1 percent (3).

Comparative Statistics for Mortality of Abdominal Wounds

A study of 384 casualties who had abdominal wounds, covering a 17-month period in Korea, was compared with patients having similar injuries from the Second Auxilary Surgical Group in World War II (4). This number includes the 75 casualties discussed at the beginning of this report. A comparison, by organs, shows that the colon, jejunum, ileum and liver were the most commonly injured during both the Korean conflict and World War II (table 6). The stomach, kidney and spleen were the second most commonly injured organs. In those instances in which injury occurred to more than one organ, the mortality rate in Korea was approximately one-half of the mortality rate in World War II. The mortality rate was approximately the same for injuries to the duodenum and the ureter.

Mortality of Thoraco-Abdominal Wounds by Side of Body Affected

There were 72 casualties in the Korean conflict who had thoraco-abdominal wounds with perforation of the right side of the diaphragm (table 7). Twelve of these patients died, a mortality rate of 16.6 percent. The Second Auxiliary Surgical Group of World War II reported 103 deaths among 435 casualties, a mortality rate of 24 per-


429

Table 6. Comparative Statistics of Mortality Rates for Abdominal Wounds by Organs-Korea, 1952-53, and World
War II, 1942-45*


Viscus

Korea

World War II

All cases, deaths

Complicated, percent dying

Uncomplicated, percent
dying

All cases,* deaths

Complicated, percent dying

Uncomplicated, percent dying

Number

Percent

Number

Percent

Colon

140

15. 0

18. 5

9. 3

1,106

37

41

23

Jejunum and Ileum

134

13. 4

16. 4

3. 0

1,168

30

36

14

Liver

102

15. 6

20. 6

9. 0

829

27

39

10

Stomach

45

17. 5

22. 8

0

416

41

42

29

Kidney

55

25. 4

29. 1

0

427

35

38

16

Spleen

54

15. 0

18. 6

0

341

25

30

12

Rectum

22

18. 1

16. 0

11. 7

155

30

42

14

Bladder

21

9. 4

13. 3

0

155

30

34

0

Duodenum

17

41. 1

43. 7

0

118

56

56

50

Pancreas

9

22. 2

25. 0

0

62

58

57

100

Gallbladder

33

0

0

0

53

30

30

0

Ureter

4

50. 0

50. 0

0

27

41

42

0

*Second Auxiliary Surgical Group from Beebe and DeBakey

cent (4). In 57 patients, the left side of the diaphragm was perforated; a mortality rate of 8.7 percent as compared with a mortality rate of 30 percent in World War II. The mortality rate of all wounds of both sides of the diaphragm was 13.1 percent for Korea as compared with 27 percent in World War II. In Korea, wounds of the right side of the diaphragm carried a higher mortality rate than those of the left side, probably because of injury to the liver. In World War II, there was a mortality rate of 24 percent for wounds of the right side and 30 percent for wounds of the left side (4).

Table 7. Comparative Statistics of Mortality Rates for Thoraco-abdominal Wounds by Side of Body Affected

Theater

Right diaphragm

Left diaphragm

Total

Number wounded

Deaths

Number wounded

Deaths

Number wounded

Deaths

Number

Percent

Number

Percent

Number

Percent†

Korea, 1952-53

72

12

16. 6

57

5

8. 7

129

17

13. 1

World War II,*
1942-45


435


103


24. 0


448


136


30 .0


883


239


27. 0

*Second Auxiliary Surgical Group from Beebe and DeBakey.
†Chi Square Test=10.8; P=0.001.


430

Comparative Statistics for Abdominal Wounds by Number of Organs Damaged

A direct comparison of the number of organs damaged was made between those casualties with abdominal wounds in Korea and those from World War II (table 8) (4). It was felt that this would be an accurate comparison, since it was based on the number of organs injured. There was little difference in the time interval from injury to operation: in World War II, from zero to 7 hours; in Korea, an average of 7 hours.

Table 8. Comparative Statistics of Mortality Rate for Abdominal Wounds By Number of Organs Damaged


Number of abdominal organs hit

Korea-Injury to operation, average 7 hours

World War II-Injury to operation,* 7 hours or less


Statistical tests for significance of difference

Number wounded

Percentage dying

Number wounded

Percentage dying

0

36

2. 8

98

5


Chi Square analyzed according to number of abdominal organs injured gives:

X2 (5) = 11.95
0.05 P 0.02

1

181

6. 62

496

10

2

102

6. 82

402

24

3

45

26. 6

132

42

4

12

8. 3

41

54

5

6

16. 0

13

92

6

2

50. 0

3

100


Total


384


10. 68


1,185


20. 51

Overall statistics:

X
S.D.=4.35
  P 0.0003

*Second Auxiliary Surgical Group from Beebe and DeBakey.

The casualties without organ damage had essentially the same mortality rate. Little difference was seen in the mortality rates of casualties who had four, five, or six organs damaged. There was a significant difference between the overall mortality of 10.68 percent in Korea and 20.51 percent in World War II. Although there was little difference in the slightly wounded (no organ damage) and the most severely wounded (four, five, or six organ damage), the significant differences in the moderately wounded (one, two, or three organ damage) and the overall mortality rate points out the improved results achieved in Korea.

Summary

A study of the casualties who had abdominal and thoraco-abdominal wounds was made at the 46th Surgical Hospital in Korea.


431

The surgical technics proven to be acceptable in the management of these casualties have been outlined.

The average time from injury to admission for casualties with abdominal wounds was 3.1 hours; for thoraco-abdominal wounds, 4.7 hours. The abdominal casualties received an average of 3,400 cc. of blood and the thoraco-abdominal casualties received an average of 2,800 cc. of blood in the first 24 hours after injury. More than half of the deaths were caused by uncontrolled hemorrhage.

The mortality rate for 129 thoraco-abdominal casualties in the last year and a half of the Korean war was 13.1 percent. In World War II the mortality rate for 883 casualties with thoraco-abdominal wounds was 27 percent.

The mortality rate for 384 casualties who had abdominal injuries in Korea was 10.68 percent. In a series of 1,185 casualties who had comparable organ damage and comparable evacuation times treated by the Second Auxiliary Surgical Group, the mortality rate was 20.51 percent.

Acknowledgment

Some of the data pertaining to the Korean conflict in tables 6, 7, and 8 were compiled from the records of Captain John Howard and Captain Frank Inui.

References

1. Holmes, R. H.: Regional Distribution of Wounds. Personal communication.

2. Welch, C. S., Tuhy, J. E., and Pearson, R. W.: Abdominal Surgery in the Evacuation Hospital. Surgery 21: 1, 1947.

3. Chest and Abdomen, War Supplement No. 3. British Journal of Surgery.

4. Beebe, G. W., and DeBakey, M. E.: Battle Casualties. Charles C. Thomas, Springfield, Illinois, 1952.