The American Journal of Surgery (2013) 206, 613-618
Editorial Opinion
Did James A. Garfield die of cholecystitis? Revisiting the
autopsy of the 20th president of the United States
Theodore N. Pappas, M.D.a, Shahrzad Joharifard, M.D.b,*
a
Department of Surgery, Duke University, Durham, NC, USA; bDepartment of Surgery, University of British Columbia,
Faculty of Medicine, 910 West 10th Avenue, Vancouver, British Columbia, Canada V5Z 4E3
Historians have reviewed the shooting and ultimate
demise of James A. Garfield since 1881, the year of his
assassination. Remarkably, rather than faulting the assassin,
nearly every author has blamed the President’s death on the
physicians who cared for him in the 80 days after the
shooting. Most historians have argued that Garfield’s physicians mismanaged the bullet wound on his back, leading to
iatrogenic starvation, uncontrolled sepsis, and finally death.
This article reconstructs the events after the shooting and
revisits Garfield’s postinjury clinical course in light of the
anatomic facts revealed by his autopsy. Based on this
reconstruction, we argue that the President’s medical team
followed the standard of care of the era. Furthermore, our
detailed analysis of Garfield’s postmortem examination
leads us to contend that the President died not as a result of
his physicians’ alleged incompetence but rather from a late
rupture of a splenic artery pseudoaneurysm after suffering
from a complicated course of acute cholecystitis.
entering the station, Garfield was approached from behind
by a mentally unstable attorney, Charles J. Guiteau, who
twice fired his 0.44-caliber revolver at close range.1 The
first shot grazed the President’s right arm, whereas the second entered his back between the 11th and 12th ribs, four
inches to the right of the vertebral column. The President
immediately fell to the ground, ‘‘blood streaming from
his two wounds.’’2 Meanwhile, Blaine and two local police
officers tackled Guiteau as he attempted to escape through a
nearby exit.2
Within minutes, a throng of people surrounded the
President. The first physician on the scene was the District
Health Officer, Dr. Smith Townsend, who discovered Garfield ‘‘apparently dying,’’ the floor ‘‘covered in blood.’’2 The
President remained conscious, immediately complaining of
Assassination Attempt
On the morning of July 2, 1881, James A. Garfield was
en route to his 25th reunion at Williams College. The
President was accompanied by the Secretary of State,
James G. Blaine, with plans to board the 9 AM train for
Massachusetts from the Baltimore and Potomac Railroad
Station in downtown Washington, DC. Shortly after
The authors declare no conflicts of interest.
* Corresponding author. Tel.: 11-604-356-9953; fax: 11-604-8754036.
E-mail address: sjoharifard@gmail.com
Manuscript received August 23, 2012; revised manuscript December
10, 2012
0002-9610/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.amjsurg.2013.02.007
James A. Garfield, circa 1881.
614
severe pain in his back and weakness in his legs.3 Based on
the autopsy findings, this was likely related to the bullet traversing the L1 vertebral body.4 The path of the bullet through
L1 would not necessarily have resulted in paralysis but certainly could have created edema around the spinal cord,
leading to the neurologic symptoms reported.
Dr. Townsend ordered Garfield to be moved to the second
floor of the train station on a mattress, where Drs. D.W. Bliss
and Robert Reyburn, two prominent local physicians, ultimately received him. Dr. Bliss, finding the patient ‘‘deathly
pale, almost pulseless and apparently dying of internal
hemorrhage,’’ proceeded to examine the flank wound, first
with his finger and then with a Nélaton probe.2,5 Although
this unsterile examination of the wound may be surprising
to today’s readers, probing wounds was considered the standard of care for gunshot injuries in the 19th century. Given
that bullet wounds are all contaminated by pieces of clothing
and other foreign bodies, the use of a finger or probe to examine these wounds was thought to be of great benefit and
little harm in the 1880s.6 The bullet was often a nidus for infection, and, thus, early removal was generally considered
therapeutic. Furthermore, at a time when x-ray technology
was in its infancy, fingers and probes were the sole tools at
the physician’s disposal to define the course of the bullet
and, therefore, the patient’s prognosis.
Based on their initial probing of the wound, Garfield’s
physicians assumed that the second bullet had lodged in the
peritoneal cavity, possibly after passing through the liver.
Performing an exploratory laparotomy in the case of
abdominal trauma was an extremely controversial notion
in 1881; in fact, the procedure was not widely accepted as a
viable treatment option for penetrating abdominal trauma
until the turn of the century.7 As such, Garfield’s physicians
had little option but to follow the standard of care of the era
and treat the President with nonoperative management.
Thus, Garfield was moved to the White House via ambulance, where he was fully expected to die overnight. The
President’s staff sent a telegraph to summon Mrs. Garfield,
who was recovering from a bout of malaria on the New Jersey coast.8 Upon arriving at the White House, the First
Lady was given a private audience with her husband to
say her goodbyes.9
To the surprise of his doctors, however, Garfield
survived the night and began a slow recovery. As the
autopsy showed, after traversing L1, the bullet injured the
splenic artery near its takeoff from the celiac axis.4 Remarkably, the President did not bleed to death despite
this life-threatening arterial injury. Instead, a pseudoaneurysm of the splenic artery formed as the bleeding stopped,
aided by the tamponade effect of the retroperitoneum.4
In the ensuing 80 days, the President’s medical care
captured the attention of the nation. Garfield’s physicians
issued daily bulletins on their patient’s condition, while
newspapers and medical journals published regular commentary from physicians around the world debating the
likely course of the bullet and the President’s chance of
surviving.
The American Journal of Surgery, Vol 206, No 4, October 2013
The President’s Physicians
Garfield was cared for by some of the best trauma
surgeons of the era. David Hayes Agnew, the senior-most
consultant attending to the President, was perhaps the most
prodigious surgeon of his day, possessing extensive Civil
War experience in the treatment of gunshot wounds. He
was a Professor of Surgery at the University of Pennsylvania’s School of Medicine from 1863 to 1884 and served as
President of the American Surgical Association in 1887.10
Agnew’s most important textbook, The Principles and
Practice of Surgery (1878 to 1883), was the authoritative
surgical text of its day.11
Doctor Willard Bliss (Doctor was his first name) was
one of the most prominent physicians in the nation’s capital
in 1881. Bliss received his medical training at the
Cleveland Medical College and was considered an expert
in ballistics, also because of his considerable Civil War
experience. Additionally, he was appointed the head of
Washington, DC’s Armory Square Hospital after the war.
Interestingly, Garfield was not the first President that Bliss
attended to; he treated Zachary Taylor, the 12th President of
the United States, in 1844 for malaria.12
Robert Reyburn was also a prominent medical figure in
Washington, DC in the 1880s. He received his medical
degree from the Philadelphia College of Medicine and
Surgery in 1856. Dr. Reyburn was subsequently appointed
one of the first five faculty members at Howard University’s
School of Medicine and served as Dean from 1870 to
1871.13
Lastly, Joseph Janvier Woodward was a Civil War
surgeon who performed the autopsy of both Abraham
Lincoln and John Wilkes Booth. He consulted in the care of
Garfield and assisted in his autopsy.14
The President’s Clinical Course
In the days immediately after the shooting, Garfield’s
physicians remained preoccupied with determining the path
of the bullet. The wound was probed on multiple occasions,
but contrary to what previous historians have argued, this
was not done with ‘‘unwashed and manure-tainted
hands.’’15 Rather, as Dr. Reyburn indicates in his day-today account of the President’s illness, Garfield’s physicians
made a concerted effort to adhere to Lister’s emerging theories of antisepsis. ‘‘The most scrupulous cleanliness of the
instruments and surgical appliances was observed, and also
of the antiseptic solutions used for the daily washing out of
the wound, and every effort was made to render them as antiseptic as possible.’’2,16 Regardless, the frequent probing of
the wound yielded no concrete information about the bullet’s path. Even Alexander Graham Bell, with his prototype
metal detector, was twice unsuccessful in locating the
bullet.17
Over the next two weeks, the President’s condition
gradually improved. Acknowledging that they might never
T.N. Pappas and S. Joharifard
Did James A. Garfield die of cholecystitis?
determine the exact location of the bullet, but assuming that
it had lodged in the pelvis after traversing the peritoneum,
Garfield’s physicians focused their attention on caring for
his back wound with twice daily dressing changes. Initially,
Garfield suffered from nightly fevers and frequent emesis,
but by day 14, he was eating solid food and appeared to be
convalescing.18 However, by the beginning of the third
week, the President showed signs of a brewing infection.
Surmising that Garfield’s deterioration was caused by secondary infection of the back wound, his physicians responded in kind by debriding the wound, making counter
incisions, and inserting multiple tubes to optimize drainage.2 Surprisingly, as the appearance of the wound improved, Garfield’s overall condition worsened.
The autopsy showed that the aggressive local management of the bullet entry wound resulted in a cavity extending
to the ipsilateral anterior iliac crest. Previous authors have
suggested that this largely iatrogenic cavity was the source of
the President’s sepsis and gradual decline.19,20 In actuality,
however, there is no evidence in the postmortem examination
that the back wound was a source of sepsis. On the contrary, it
appeared to be a very large but well-drained space. There was
no infection under pressure, no residual abscess, and ‘‘very
little pus’’ per the autopsy report.4 Garfield’s physicians irrigated the wound on a daily basis with carbolic acid, a common cleansing solution in the day, via a small rubber
catheter that they inserted directly into the wound. In addition, the President’s surgeons conducted at least two formal
incision and drainage procedures to be certain that the space
was adequately drained.2 Therefore, given the historical clinical data and the autopsy results, one must conclude that although Garfield’s physicians were incorrect in assuming that
the back wound was the cause of the President’s decline, their
management effectively eliminated it as an ongoing source
of sepsis.
Three to four weeks after the shooting, the President
starting eating less, developed abdominal pain, and once
again became regularly febrile. By the sixth week, the
President was clearly septic, ultimately developing parotitis
on day 47 and pneumonia on day 72. If not the back wound,
what then was the cause of the President’s abdominal pain
and delayed sepsis? The answer lies in the postmortem
examination, which describes a bile-filled abscess involving
the gallbladder. Although the autopsy report does not
mention the presence or absence of gallstones and although
there was no histologic examination of the gallbladder, the
presence of a bile-filled abscess adherent to the gallbladder
most likely represents a complication of acute cholecystitis.
An untreated, contained perforation of the gallbladder from
cholecystitis would certainly result in a continued downhill
course associated with fever, malaise, decreased appetite,
abdominal pain, and ultimately sepsis.
Death and Autopsy
On September 17th, the President became febrile, tachycardic to 120 bpm, and complained of severe pain in his
615
anterior mediastinum. Then, late in the evening on September 19th, 80 days after being shot, Garfield suffered a
severe episode of chest pain and died shortly thereafter at
10:35 PM.2 As described in detail below, the postmortem examination showed that the President’s final cause of death
was bleeding secondary to the late rupture of a traumatic
pseudoaneurysm of the splenic artery.4 The chest pain experienced two days earlier likely represents a sentinel hemorrhage before the final rupture of the pseudoaneurysm;
although the sentinel hemorrhage may have initially been
contained in the retroperitoneum, the autopsy shows that
rupture of the pseudoaneurysm eventually resulted in free
intraperitoneal hemorrhage and death.4
The postmortem examination was conducted 18 hours
after the President’s death by Drs. Bliss, Reyburn, Agnew,
and Woodward, in addition to other members of the
medical team.4 On external examination, the physicians
commented on the President’s severely emaciated state
and his extensive peripheral wasting. They also noted an ulceration behind his right ear in addition to multiple purpuric macules and pus-filled pustules covering the deceased’s
back, shoulders, axillae, and buttocks.
Early in the course of the autopsy, a flexible catheter was
inserted into the back wound, the purpose of which was to
define the bullet tract from the outside. Garfield’s physicians assumed that when they exposed the abdomen the
catheter would be visible in the peritoneal cavity, most
likely in the pelvis; however, when the abdominal contents
were explored, it became obvious that the bullet had never
entered the peritoneum. Rather than finding the bullet upon
exposure of the abdomen, Garfield’s physicians discovered
extensive adhesions, particularly between the transverse
colon and the anterior edge of the liver. In the left upper
quadrant, they discovered a ‘‘mass of black, coagulated
blood that covered and concealed the spleen and the left
margin of the greater omentum.’’4 When they pulled back
the omentum, they saw that the blood tracked down into
the pelvis where ‘‘more than a pint of bloody fluid’’ was
found.4
At this juncture, the medical team concluded that the
President died from secondary hemorrhage, the source of
which was yet undiscovered. Only after removing the
abdominal organs and engaging in detailed dissection did
the examiners find ‘‘a rent, nearly four-tenths of an inch
long in the main trunk of the splenic artery, two inches and
a half to the left of the coeliac axis.’’4 Together these findings led Garfield’s physicians to determine that the President’s proximate cause of death was a late rupture of a
traumatic splenic artery pseudoaneurysm. ‘‘This hemorrhage,’’ they wrote in a bulletin that was reproduced in
The New York Times, ‘‘is believed to have been the cause
of the severe pain in the lower part of the chest complained
of just before death.’’4
A good part of the remainder of the autopsy was focused
on determining the path of the bullet. Tracing the catheter
inserted before making the initial incision, Garfield’s
physicians found an iatrogenic tract leading down from
616
The American Journal of Surgery, Vol 206, No 4, October 2013
the wound, behind the right kidney, along the iliac fascia,
and almost into the right groin. On the other hand, the
original bullet tract was ‘‘completely obliterated by the
healing process’’ and therefore had to be dissected in order
to decipher the ultimate path of the bullet.4 Much to their
surprise, Garfield’s physicians soon discovered that the bullet was encysted below the pancreas, not in the pelvis as
they had thought:
It was found that from the point at which it had fractured
the right eleventh rib (three inches and a half to the right
of the vertebral spines) the missile had gone to the left
obliquely forward, passing through the body of the first
lumbar vertebra, and lodging in the adipose connective
tissue immediately below the lower border of the
pancreas, about two inches and a half to the left of the
spinal column and behind the peritoneum.4
After discovering that the bullet tract was ‘‘almost free
of pus,’’ and thus an unlikely source of infection, Garfield’s
medical team was forced to look elsewhere for the source
of the President’s septic condition.4 Although there was
certainly evidence of a congestive pneumonia in the right
lung, Garfield did not complain of cough until very late
in his postinjury clinical course. Moreover, the parotitis
and disseminated pustules appeared after the commencement of fever and rigors. Upon completing the autopsy,
Garfield’s physicians opined that the ‘‘fractured, spongy tissue of the vertebrae furnish[ed] a sufficient explanation of
the septic condition which existed.’’4 In arriving at this conclusion, however, the President’s medical team overlooked
a major unexpected finding of the autopsydthe discovery
of a large abdominal abscess. The original description of
this finding merits inclusion in its entirety:
The abdominal viscera was then carefully removed from
the body, placed in suitable vessels, and examined
seriatim, with the following result: The adhesions
between the liver and the transverse colon proved to
be [be] bound [by an] abscess cavity between the
undersurface of the liver, the transverse colon, and the
transverse [mesocolon], which involved the gallbladder,
and extended to about the same distance on each side of
it, measuring 6 inches transversely and 4 inches from
before backward. This cavity was lined by a thick
pyogenic membrane, which completely replaced the
capsule of that part of the under surface of the liver
occupied by the abscess. It contained about two ounces
of greenish yellow fluidda mixture of pus and biliary
matter. This abscess did not involve any portion of the
substance of the liver except the surface with which it
was in contact, and no communication could be detected
between it and any part of the wound.4
The location of the abscess between the liver and the
transverse colon, combined with the fact that it was reportedly filled with ‘‘greenish yellow fluid,’’ points toward a
perforation of the gallbladder.4 An alternate explanation of
the pericholecystic fluid collection is that it was an abdominal abscess caused by hematologic seeding of the peritoneal
cavity from the infected back wound. This interpretation,
however, does not explain the presence of bile in the abscess.
Furthermore, the autopsy report states that the bullet did not
pass anywhere near the liver or gallbladder, while later
clearly indicating that there was no connection between
the abscess and the back wound. Therefore, the only explanation that seems probable is that the abscess formed because of the perforation of the President’s gallbladder three
to four weeks after he was shot. Moreover, this gallbladder
abscess was the most logical source of the President’s sepsis,
which was, in turn, the cause of Garfield’s unrelenting downhill course.
Why Did Garfield’s Physicians Miss the
Significance of the Gallbladder-Associated
Abscess?
The simplest answer is that Garfield’s physicians may not
have previously seen autopsy evidence of cholecystitis, nor
would they have expected to discover this pathology in an
abdominal gunshot victim. Although reports of cholecystitis
after the treatment of non-gallbladder disease first emerged
in 1844, the disease process was not well characterized until
Frank Glenn’s treatise on the subject presented at the
American Surgical Association in 1947.21–23 Furthermore,
the association between post-traumatic conditions and acalculous cholecystitis was not well documented until 1970,
when Lindberg and Grinnan24 reported their experience
with 12 American Vietnam War soldiers who developed cholecystitis within 10 to 35 days after the management of their
war injuries. Given the somewhat recent delineation of the
pathophysiology behind the development of cholecystitis
in the context of treatment of non-gallbladder disease, it is
not surprising that Garfield’s physicians not only failed to
recognize that the President had developed cholecystitis during his recovery from the shooting but also failed to recognize the significance of the gallbladder abscess they
encountered during the postmortem examination.
Why Did Garfield’s Historians Miss the
Significance of the Gallbladder-Associated
Abscess?
The reason why historians have failed to recognize the
significance of the four by six inch abscess discovered
during the President’s postmortem examination is harder to
delineate. Perhaps the failure of Garfield’s medical team to
recognize the importance of this critical autopsy finding has
led historians to similarly choose to look beyond the
obvious. In addition, it has always been convenient to
accuse the President’s doctors of malpractice, suggesting
that Garfield’s demise was secondary to iatrogenic sepsis
from repeated probing of the bullet wound. This allegation
is made ever so more enticing given that it was mimed by
Guiteau at his trial; when asked why he killed the President,
T.N. Pappas and S. Joharifard
Did James A. Garfield die of cholecystitis?
the assassin replied, ‘‘I simply shot him, it was the doctors
that killed him.’’25
Everyone from Alexander Graham Bell to David
Hayes Agnew had an opinion about the President’s
declining health during his postshooting clinical course.
There was speculation from a variety of sources that
Garfield’s physicians were at best in disagreement and at
worst incompetent. One of the main conclusions of the
autopsydnamely the fact that the President’s doctors had
completely misjudged the bullet tractdseemed to support
the contention that Garfield’s medical team had been
mistaken in their approach. Although the presumption of
intraperitoneal injury certainly colored much of the
medical team’s thinking and care, it is not clear that
Garfield’s physicians would have managed the case
differently had they known that the bullet tracked toward
the President’s left side extraperitoneally through the L1
vertebra. As the pre-eminent surgeon J. Marion Sims so
eloquently argued after Garfield’s death, the President’s
physicians adhered strictly to the standard of care of
the time:
But if the course of the ball could have been ascertained,
the result would have been precisely the same. The
surgeons could only have kept the pus pouch between
the fractured ribs and spine, and the one extending down
into the pelvis, empty and clean. This they did all the
time as the history of the case and the autopsy show.26
Furthermore, even if Garfield’s physicians had possessed the means to ascertain that the President had
developed an abscess around his gallbladder, it seems
highly unlikely that they could have intervened. The first
reports of a cholecystotomy and cholecystostomy had
only recently been published in 1866 and 1878, respectively.27–29 The first successful cholecystectomy, moreover, was not reported by Langenbuch until the year
after the President’s death.30,31
Comments
Most contemporary authors have chosen to take the
obvious errors in diagnosis that were unveiled by the
autopsy as proof that Garfield’s physicians committed gross
negligence.19,20 Although the postmortem examination
made it abundantly clear that the President’s proximate
cause of death was a ruptured pseudoaneurysm, many historians are happy to continue to place all responsibility on
Garfield’s physicians. Why some historians have been unwilling to distribute the responsibility for the President’s
death at least partially on the assassin is unclear; perhaps
it makes for better historical reading to blame the doctors
and not the individual who was holding the gun.
Still, one cannot ignore the unexpected discovery of
a gallbladder-related abscess. Placing this autopsy
finding in context with the patient’s overall clinical
course leads the reader to two plausible conclusions.
First, President Garfield died of a ruptured
617
pseudoaneurysm of the splenic artery, a complication
of the initial gunshot wound. Second, the President’s
clinical course, highlighted by recurrent episodes of
anorexia, extreme weight loss, recurrent fevers, abdominal pain, and small subcutaneous abscesses, was not
caused by an infected bullet tract but was more likely
secondary to complications of an undiagnosed and
unmanaged course of cholecystitis.
References
1. Mr. Blaine’s story of it: what the Premier saw of the assassination of
Mr. Garfield. Washington Post; July 3, 1881:1.
2. Reyburn R. Clinical history of the case of President James Abram Garfield. JAMA 1894;XXII: 411–7; 460–4; 498–502; 545–9; 578–82;
621–4; 664–9.
3. Hamilton FH. A Review of the late President Garfield’s Case. Br Med
J 1881;2:619–20.
4. Verdict of the surgeons: the autopsy on the remains of President Garfield. New York Times; October 2, 1881:1–1.
5. Nélaton M. New methods of discovering the presence of a ball or other
metallic body within a wound-Nelaton’s porcelain probe-Favre’s galvanic probe. Am J Med Sci 1863;45:499.
6. Longmore T. Remarks on the Instruments Designed for Exploring
Gun-Shot Wounds, with a View to Detect Bullets or Other Foreign Bodies Suspected to be Lodged in them. Br Med J 1871;
2:751–2.
7. Nancrede CB. I. Should Laparotomy be Done for Penetrating Gunshot
Wounds of the Abdomen, Involving he Viscera. Ann Surgery 1887;5:
465–90.
8. Sending for Mrs. Garfield: receiving the sad news with fortitude. New
York Times; July 3, 1881:7.
9. Condition of the President at one o’clock Sunday morning. Baltimore
Sun; July 4, 1881:5.
10. Dr. Hayes Agnew dead: the great surgeon yields to an attack of bronchitis. Washington Post; March 23, 1892:4.
11. Agnew DH. The Principles and Practice of Surgery. J.B. Lippincott
and Co., Philadelphia, USA. 1883.
12. Garfield’s doctor dead. Chicago Daily Tribune; February 22,
1889:3.
13. Dr. Reyburn’s death puts Howard in gloom. Baltimore Afro-American;
April 3, 1909:1.
14. Death of surgeon Woodward. Baltimore Sun; August 19, 1884:1.
15. Herr HW. Ignorance is bliss: the Listerian revolution and education of
American surgeons. J Urol 2007;177:457–60.
16. Lister JB. Antiseptic Principle in the Practice of Surgery. Br Med J
September 21, 1867.
17. Yesterday: the bullet finder. Chicago Tribune; August 2, 1881:1.
18. The White House patient: President Garfield’s continued improvement. New York Times; July 16, 1881.
19. Rutkow IM. James A. Garfield. New York, New York, USA: Macmillan; 2006.
20. Millard C. Destiny of the Republic. New York, New York, USA: Doubleday; 2011.
21. Duncan J. Femoral hernial; gangrene of the gallbladder; extravasation
of bile; peritonitis; death. North J Med 1844;2:1151–3.
22. Glenn F. Acute cholecystitis following the surgical treatment of unrelated disease. Ann Surg 1947;126:4.
23. Glenn F, Moore S. Gangrene and perforation of the wall of
the gallbladder: a sequela of acute cholecystitis. Arch Surg
1942;44:4.
24. Lindberg EF, Grinnan GL. Acalculous cholecystitis in Viet Nam casualties. Ann Surg 1970;17:152–7.
25. Assassin and avenger: the trial of Guiteau and his alleged assailant.
Washington Post; November 21, 1881:1.
618
26. Hammond WA, Ashhurst J, Sims JM, et al. The surgical treatment of
President Garfield. North Am Rev 1881;133:578–610.
27. Bobbs JS. Case of lithotomy of the gallbladder. Trans Indiana Med Soc
1866;18:68.
28. Alvarez WCW. Dr. Bobbs’ first operation for drainage of a diseased
gallbladder. Gastroenterology 1970;58:128–9.
The American Journal of Surgery, Vol 206, No 4, October 2013
29. Sims JM. Remarks on cholecystotomy in dropsy of the gall-bladder.
BMJ 1878;1:811–5.
30. Langenbach CJA. Ein Fall von Exstirpation der Gallenblasewegen Chronischer Cholelithiasip Heilung. Berliner Klin Wochenschr 1882;19:725–7.
31. Traverso LW. Carl Langenbuch and the first cholecystectomy. Am J
Surg 1976;132:81–2.