SCIENTIFIC ARTICLE

Proximal Pole Scaphoid Fractures: A Computed

Tomographic Assessment of Outcomes
Ruby Grewal, MD, MSc,* Kristina Lutz, MD,* Joy C. MacDermid, MSc, PhD,† Nina Suh, MD*

Purpose To report on union rates and times for a cohort of acute nondisplaced or minimally
displaced proximal pole fractures evaluated with serial computed tomography (CT) scans.
Methods All patients with isolated acute proximal pole scaphoid fractures (< 6 weeks from
injury) who presented at our institution between 2006 and 2013 were identified. Each subject’s
health record, CT scans (performed at initial assessment and serially to document healing), and
x-rays were retrospectively reviewed to determine details of injury, treatment course, and
treatment outcome. Union incidence and time to union were determined based on CT scan
results. The effect that each predictor variable had on union, nonunion, and delayed union was
assessed.
Results This cohort consisted of 53 patients with proximal pole scaphoid fractures (47 males
and 6 females; mean age, 30
17 years). The overall union incidence with cast treatment was
90% (47 of 52). The study was underpowered to detect any factors that were predictive of
developing a nonunion with cast treatment with the exception of a slight delay to seeking
treatment. Average time to union was 14
8 weeks for cases treated with surgical fixation (n ¼
4; cases that failed casting and were subsequently treated surgically) and 14
12 weeks for
cases treated with casting alone. Factors found to be correlated to longer union times included
fracture translation (r ¼ 0.30) and the presence of cysts or comminution.
Conclusions The reported union incidence and union times in this study compared favorably with
the literature. Risk factors that were associated with a significantly greater time to union
included fracture comminution, the presence of cysts, and fracture translation. Our sample size
was relatively small, and other limitations inherent in the retrospective design must be
considered. (J Hand Surg Am. 2016;41(1):54e58. Copyright Ó 2016 by the American Society
for Surgery of the Hand. All rights reserved.)
Type of study/level of evidence Prognostic IV.
Key words Computed tomography, proximal pole, scaphoid fracture, union, outcomes.

fractures. Their reported incidence varies between 6%3

P
ROXIMAL POLE SCAPHOID FRACTUREShave the worst
prognosis of any scaphoid fracture.1 However, a
review of the literature reveals few reports that
focus specifically on examining their outcomes, partic-
ularly with cast treatment.2 One reason for this is that
they occur much less frequently than waist or distal pole
and 20%.4 In fact, some of the largest series reporting
outcomes of scaphoid fractures have few proximal pole
fractures represented, and those that do focus on prox-
imal pole fractures have small sample sizes (n ¼ 4
proximal pole to 100 scaphoid fractures5; n ¼7 proximal

From the *Division of Orthopedic Surgery, and the †Clinical Research Laboratory, Division of
Orthopedic Surgery, University of Western Ontario, Roth j McFarlane Hand and Upper Limb
Center, St. Joseph’s Health Care, London, Ontario, Canada.
Received for publication July 7, 2015; accepted in revised form October 13, 2015.
No benefits in any form have been received or will be received related directly or indirectly
to the subject of this article.
Corresponding author: Ruby Grewal, MD, MSc, Division of Orthopedic Surgery. University
of Western Ontario, Roth j McFarlane Hand and Upper Limb Center, St. Joseph’s Health
Care, 268 Grosvenor St., London, Ontario N6A 4L6, Canada; e-mail: rgrewa@uwo.ca.
0363-5023/16/4101-0009$36.00/0
http://dx.doi.org/10.1016/j.jhsa.2015.10.013

54 r Ó 2016 ASSH r Published by Elsevier, Inc. All rights reserved.

 OUTCOMES OF PROXIMAL POLE SCAPHOID FRACTURES
pole to 323 scaphoid fractures6; n ¼ 22 proximal pole to
248 scaphoid fractures3).
Achieving union can be challenging for proximal
pole scaphoid fractures owing to the small fragment
size and poor vascularity. The literature reports varied
union incidences and times with casting. Nonunion
incidence varies between 5% and 50%.3,5,7 Union times
also vary. Most report union times for cast treatment
average 3 to 6 months,7e12 but 1 report indicated union
times as high as 9 months.13 These numbers have
predominantly been extracted from review papers with
limited hard data supporting these values.7e9 The pur-
pose of this study was to report on union incidence
and union times for a cohort of acute proximal pole
fractures (< 6 weeks from injury) that were non-
displaced or minimally displaced ( 1mm displace-
ment) and were evaluated with serial computed
tomography (CT) scans.
METHODS
After obtaining approval by our institution’s review
board, a departmental radiology CT database was
searched for all scaphoid CT scans conducted between
January 2006 and December 2013 inclusive. Because
this project involved primarily a chart review, in-
formed consent from each patient was not required by
our institutional review board. All scaphoids were
scanned using a CT scanning technique previously
reported.14 All proximal pole scaphoid fractures were
identified, and each subject’s x-rays, CT scans, and
health record were reviewed to determine the details
of injury, treatment course, and treatment outcome.
A total of 531 scaphoid fractures were identified, of
which 87 involved the proximal pole (16%). Fractures
presenting greater than 6 weeks from initial injury (n ¼
27) and those that were loss to follow-up before union
or nonunion could be confirmed (n ¼ 5) were excluded.
Fractures with angular deformity in the sagittal plane
(n ¼ 1) or displacement greater than 1 mm (n ¼ 1) were
also excluded. This resulted in 53 subjects for inclu-
sion in this study.
All CT scans were reviewed by 1 of 2 observers
(a senior orthopedic resident [K.L.] and hand surgery
consultant [R.G.]). The percentage of bone bridging
for each scan was determined based on the method
described by Singh et al.15 If bridging was 50% or
greater, we considered the fracture to be united and
determined the time from injury to union.16 If at least
50% bridging was not achieved and cast treatment
was abandoned as per the discretion of the treating
surgeon, cast treatment was deemed a failure and
surgery was offered. Union times for both surgically
J Hand Surg Am.
55
TABLE 1. Description of Cohort
Mean Age (y) 30
17 (range, 13e80)
Sex 47 males 6 females
Hand dominance 23 nondominant 20 dominant
hand affected hand affected
(10 unknown)
Mean time between injury 6
11 (range, 0e35)
and treatment (d)
Smoking 8 smokers 33 nonsmokers
(12 unknown)
Comorbidities 2 diabetics
and nonsurgically treated fractures were calculated
separately. There was 1 case in which the nonunion
was accepted after 3 months of casting. Because the
patient was 80 years old and asymptomatic, no
additional treatment was offered. Because union was
not achieved or attempted, this case was excluded
from the time to union calculations. Additional de-
tails such as orientation of fracture line, translation of
fracture fragments, presence of a humpback deformity,
presence of comminution, features associated with
avascular necrosis, cysts, and/or sclerosis along the
fracture line were assessed on the CT scan.
Statistical methods
The effect that each predictor variable had on union,
nonunion, and time to union was assessed. Categor-
ical variables were assessed with a chi-square test
(Fisher exact where appropriate) and continuous
variables were compared with a t test or an analysis of
variance test where appropriate.
RESULTS
Description of cohort
This cohort consisted of 53 patients with proximal
pole scaphoid fractures. Descriptive details are outlined
in Tables 1 and 2. All fractures were given a trial of cast
immobilization (short-arm thumb spica cast [SATSC])
except for 1 patient who was treated with open reduc-
tion internal fixation (ORIF) after the initial assessment
because he was an elite athlete who was interested in
optimizing his chances for an early recovery (non-
displaced fracture; union time, 6 weeks).
Nonunion: The overall union incidence for cast treat-
ment alone in this cohort was 90% (47 of 52). The study
was underpowered to detect any factors that were
predictive of developing a nonunion with cast treat-
ment (Table 3), other than a slight delay to seeking
treatment (P ¼ .03). Fractures that were successfully
r Vol. 41, January 2016
56 OUTCOMES OF PROXIMAL POLE SCAPHOID FRACTURES

TABLE 2. Description of Fractures

Fracture orientation 40 transverse (76%) 12 oblique 1 avulsion
Translation of fracture fragments 8 displaced Range, 0e1 mm Mean translation, 0.11
0.29 mm
Cysts 10 cysts 43 no cysts
Comminution 6 comminuted 47 no comminution 2 fragmentation of proximal pole
Sclerosis along fracture line 3 sclerosis 50 no sclerosis

TABLE 3. Potential Risk Factors and Association TABLE 4. Union Times Based on Presence and
With Nonunion Absence of Suspected Risk Factors*
Age P ¼ .88 Risk Factor Present
P
Sex P ¼ .40
No Yes Value
Smoking history P ¼ .44
Fracture orientation P ¼ .40 Sex Female, 12
6 Male, 17
15 .43

Comminution P ¼ .40 Smoking 17
16 18
11 .90

Fragmentation of proximal pole P ¼ .64 Fragmentation 16
15 22
2 .60

Cysts P ¼ .22 Comminution 15
14 28
20 .03

Treatment delay* P [ .03 Cysts 14
12 27
22 .01

*Data represent mean values
SD, in weeks.

*Treatment delay for those who were successful with casting along
Bold type indicates statistical significance (P < .05).
was 5
10 days vs 17
17 days delay to treatment for those who
were not successful with casting alone.
Bold type indicates statistical significance (P < .05).

with casting (95% union as per CT); however, 1 month

with cast treatment were treated on average 5
10 days
from the time of injury versus those that were unsuc-
cessfully treated with casting who averaged 17
17
days. Although the study was underpowered to detect
a difference, displacement of minimally displaced
fractures was not a factor significantly related to the
risk of developing into a nonunion (P ¼ .88) because
the mean displacement in those that healed success-
fully with casting was 0.12
0.31 mm and in those
that did not heal successfully was 0.10
0.22 mm;
however, this cohort included only minimally dis-
placed fractures. There were 5 patients who failed to
unite with casting, 3 of whom were initially given a
trial of SATSC; however, when serial CT scans failed to
demonstrate progression of union, they were offered
surgery. The time interval between injury and the de-
cision to abandon cast treatment was dependent on
the treating surgeon’s discretion (6 weeks, 10 weeks,
and 5 months). All 3 patients successfully united their
fractures after ORIF (1 iliac crest bone graft, 1 cancel-
lous distal radius bone graft, 1 without bone graft). The
time required for union was 3.6, 5.5, and 4.0 months,
respectively, from their surgery date. The fourth patient
was an 80-year-old man for whom no further treatment
was offered, and the nonunion was accepted after 3
months of casting had failed.
The fifth subject who failed casting had a non-
displaced proximal pole fracture that initially healed
later, he sustained another fall. A repeat CT scan de-
monstrated 75% bone bridging so the patient was not
recasted. Six months later, he was rescanned because of
persistent pain, and the CT demonstrated 10% bridging
at the fracture site. He was casted for a month; however,
bridging decreased to 5% and the proximal pole
developed signs consistent with avascular necrosis. He
then received a vascularized bone graft with ORIF and
bone bridging was 70% after 5.5 weeks. A CT scan 6
months later showed that bridging had increased to
90%, and he was asymptomatic at final follow-up.
Time to union and delayed union: The mean time from
surgery to union was 14
8 weeks, and the mean time
to union for the patients who were treated with casting
alone was 14
12 weeks. Age (r ¼ e0.26; P ¼ .058)
and fracture translation (r ¼ 0.30; P ¼ .03) were found
to be weakly correlated with union time. Additional
factors associated with a significant increase in union
time are listed in Table 4. When cases with cystic
changes and comminution were excluded, the mean
union time was 11
7 weeks (n ¼ 38).
DISCUSSION
Proximal pole scaphoid fractures are relatively un-
common injuries with limited reports of their out-
comes. A recent meta-analysis evaluating nonsurgical
treatment of acute scaphoid fractures identified 1,147
fractures, and 67 (6%) involved the proximal pole.2

J Hand Surg Am. r Vol. 41, January 2016

 OUTCOMES OF PROXIMAL POLE SCAPHOID FRACTURES
TABLE 5. Reported Nonunion Rates Among
Studies Evaluating Outcomes of Proximal Pole
Fractures
Nonunion Number of Proximal
Reference Incidence (%) Pole Fractures
Eddeland et al10 15 7 needed to detect differences in rates and predictors of
17
Gellman et al 20 5 outcome. Consequently, we were underpowered to
Clay et al18 31 11
Dickison et al12 33 15
Terkelsen et al19 40 5 mean union time of 14 weeks compares favorably with
Margo & Seely 5
50 6 the existing literature (12e24 weeks). This range is
In our institution, proximal pole fractures represented
approximately 16% of all scaphoid fractures seen
over the past 8 years (87 of 531). The majority of
these fractures united (50 of 55; 91%) in a mean of 14
weeks.
Our 10% nonunion incidence is similar to both that
reported by Steinmann and Adams (5%e10%)7 and
Gholson et al (5% for nondisplaced proximal pole
fractures; n ¼ 22).3 The union incidence in our cohort
compares favorably with the reported nonunion rates
that range from 15% to 50%5,10,12,17e19 (Table 5).
Our union incidence was also higher than the inci-
dence of 66% reported in a recent meta-analysis of
nonsurgically treated proximal pole scaphoid frac-
tures that was based on some of these same historic
studies.2 Many of the studies included in that meta-
analysis did not use CT scan to evaluate union, did
not have a standardized definition for union, and had
small sample sizes. The differences could also be
related to the initial time to diagnosis and treatment
because all patients were immobilized (in either an
SATSC [51 of 53; 96%] or a removable splint [2 of
53] 4%) at the time of their initial assessment, which
was a mean 6 days from injury.
Our cohort was underpowered to identify any pre-
dictors of nonunion, including both clinical predictors
(age, sex, smoking history) and CT-based radiographic
predictors (fracture orientation, comminution, frag-
mentation, cysts, and translation). Although the study
was underpowered to detect a difference, displacement
of minimally displaced fractures was not a factor sig-
nificantly related to the risk of developing a nonunion.
This may be partially explained by the fact that there
were few displaced fractures in this cohort, and of those
that were displaced, they were only minimally dis-
placed. Some report that all proximal pole fractures are
considered unstable, even those that are nondis-
placed.20 This may explain our observations, in that
perhaps the additional instability induced by minimal
J Hand Surg Am.
57
displacement did not affect union sufficiently to be
detected in this series, although translation of fragments
was weakly correlated to union times. Our event rate (5
nonunions) and sample size (n ¼ 53) were both rela-
tively small. When each patient is classified as united or
nonunited and proportions are low, large samples are
detect the influences of these factors on nonunion.
Union time was evaluated using CT scan, and the
derived from several review articles providing a gen-
eral range of expected union times (3e6 months) and
not from actual data.7e9 The studies that include raw
data reporting union times were based on x-ray find-
ings and had small sample sizes. Gellman et al17 re-
ported a mean union time of 15 weeks for patients
treated in an SATSC (n ¼ 2) and 9 weeks for patients
treated in a long-arm thumb spica cast (n ¼ 2). Edde-
land et al10 reported a mean union time of 19 weeks
(n ¼ 6), Dickison and Shannon12 reported a mean
union time of 20 weeks (n ¼ 10), and Stewart6 reported
a mean union time of 23 weeks (n ¼ 7).
Our union times are shorter than reported in the
literature and this likely relates to the added visualiza-
tion offered by CT. A recent series evaluating the nat-
ural history of scaphoid fractures in a cohort of children
and adolescents reported similar mean union times of
15 weeks (n ¼ 21) as our cohort.3 However, because
ours was a retrospective study, CTs were not scheduled
to optimize detection of early union. As such, our study
reports a conservative estimation of union times. Future
studies focusing on the outcomes of proximal pole
fractures should consider scheduling the scans earlier
because perhaps union is occurring even sooner than
we recognized.
We were able to identify a few risk factors that were
associated with a significantly greater time to union
(fracture comminution, cystic resorption, and trans-
lation of fracture fragments). These factors were found
to increase union times in other studies evaluating
union times in scaphoid fractures21 and are more easily
identified on CT than plain radiography.
With a wide range of reported nonunion incidence
and extensive time required for casting, some authors
favor routine ORIF because they feel that all proximal
pole fractures are inherently unstable, even if non-
displaced.8,20,22 However, few reports were dedicated
to outcomes of ORIF in proximal pole fractures. Rettig
and Raskin22 reported 17 patients who had a mean
union time of 10 weeks with ORIF. In their series,
union was evaluated with CT and was found to be
r Vol. 41, January 2016
58 OUTCOMES OF PROXIMAL POLE SCAPHOID FRACTURES
approximately 11 weeks for displaced and 9.5 weeks for
the nondisplaced proximal pole fractures. Slade and
Gillon23 published a retrospective review of 126 acute
scaphoid fractures treated with percutaneous ORIF.
They assessed 65 proximal pole fractures including
those associated with perilunate fracture dislocations and
distal radius fractures. They reported an overall union
rate of 99% and stated that the majority of fractures were
healed by 12 weeks, as confirmed by CT. The outcomes
reported in that series represent overall union incidence
and times for the entire cohort and were not reported
separately for the proximal pole fractures. In a series of
16 consecutive scaphoid fractures (9 waist fractures and
7 proximal pole fractures) treated with percutaneous
fixation with an Accutrak screw (Acumed, Hillsboro,
Oregon), Slade et al24 reported a 100% union rate at an
average of 12 weeks as confirmed with CT scan.
The retrospective nature of our study has inherent
weaknesses. For example, the time interval between
CT scans was not standardized, and we were unable to
account for missing data in the health record (ie,
smoking status). We also did not have a comparative
treatment arm and, therefore, cannot comment if union
incidence or union times would have varied with either
long-arm casting or with internal fixation. Given the
results of ORIF reported by various surgeons,22e24
future studies should evaluate the outcomes of prox-
imal pole scaphoid fractures treated in a randomized
controlled trial comparing ORIF versus cast immobi-
lization to offer guidance for the management of
proximal pole scaphoid fractures.
REFERENCES
1. Segalman KA, Graham TJ. Scaphoid proximal pole fractures and
nonunions. J Am Soc Surg Hand. 2004;4(4):233e249.
2. Eastley N, Singh H, Dias JJ, Taub N. Union rates after proximal
scaphoid fractures; meta-analyses and review of available evidence.
J Hand Surg Eur Vol. 2013;38(8):888e897.
3. Gholson JJ, Bae DS, Zurakowski D, Waters PM. Scaphoid fractures
in children and adolescents: contemporary injury patterns and factors
influencing time to union. J Bone Joint Surg Am. 2011;93(13):
1210e1219.
4. Kawamura K, Chung KC. Treatment of scaphoid fractures and
nonunions. J Hand Surg Am. 2008;33(6):988e997.
J Hand Surg Am.
5. Margo MK, Seely JA. A statistical review of 100 cases of fracture
of the carpal navicular bone. Clin Orthop Relat Res. 1963;31:
102e105.
6. Stewart MJ. Fractures of the carpal navicular (scaphoid); a report of
436 cases. J Bone Joint Surg Am. 1954;36(5):998e1006.
7. Steinmann SP, Adams JE. Scaphoid fractures and nonunions: diag-
nosis and treatment. J Orthop Sci. 2006;11(4):424e431.
8. Krimmer H. Management of acute fractures and nonunions of the
proximal pole of the scaphoid. J Hand Surg Br. 2002;27(3):
245e248.
9. Raskin KB, Parisi D, Baker J, Rettig ME. Dorsal open repair of
proximal pole scaphoid fractures. Hand Clin. 2001;17(4):601e610, ix.
10. Eddeland A, Eiken O, Hellgren E, Ohlsson NM. Fractures of the
scaphoid. Scand J Plast Reconstr Surg. 1975;9(3):234e239.
11. Russe O. Fracture of the carpal navicular. Diagnosis, non-operative
treatment, and operative treatment. J Bone Joint Surg Am. 1960;42:
759e768.
12. Dickison JC, Shannon JG. Fractures of the carpal scaphoid in the
Canadian army. A review and commentary. Gynecol Obstet.
1944;79(3):225e239.
13. Puopolo SM, Rettig ME. Management of acute scaphoid fractures.
Bull Hosp Jt Dis. 2003;61(3e4):160e163.
14. Bain GI, Bennett JD, Richards RS, Slethaug GP, Roth JH. Longi-
tudinal computed tomography of the scaphoid: a new technique.
Skeletal Radiol. 1995;24(4):271e273.
15. Singh HP, Forward D, Davis TR, Dawson JS, Oni JA, Downing ND.
Partial union of acute scaphoid fractures. J Hand Surg Br.
2005;30(5):440e445.
16. Hackney LA, Dodds SD. Assessment of scaphoid fracture healing.
Curr Rev Musculoskelet Med. 2011;4(1):16e22.
17. Gellman H, Caputo RJ, Carter V, Aboulafia A, McKay M. Com-
parison of short and long thumb-spica casts for non-displaced frac-
tures of the carpal scaphoid. J Bone Joint Surg Am. 1989;71(3):
354e357.
18. Clay NR, Dias JJ, Costigan PS, Gregg PJ, Barton NJ. Need the thumb
be immobilized in scaphoid fractures? A randomised prospective
trial. J Bone Joint Surg Br. 1991;73(5):828e832.
19. Terkelsen CJ, Jepsen JM. Treatment of scaphoid fractures with a
removable cast. Acta Orthop Scand. 1988;59(4):452e453.
20. O’Brien L, Herbert T. Internal fixation of acute scaphoid fractures:
a new approach to treatment. Aust N Z J Surg. 1985;55(4):
387e389.
21. Grewal R, Frakash U, Osman S, McMurtry RY. A quantitative
definition of scaphoid union: determining the inter-rater reliability of
2 techniques. J Orthop Surg Res. 2013;8:28.
22. Rettig ME, Raskin KB. Retrograde compression screw fixation of
acute proximal pole scaphoid fractures. J Hand Surg Am. 1999;24(6):
1206e1210.
23. Slade JF III, Gillon T. Retrospective review of 234 scaphoid fractures
and nonunions treated with arthroscopy for union and complications.
Scand J Surg. 2008;97(4):280e289.
24. Slade JF III, Grauer JN, Mahoney JD. Arthroscopic reduction and
percutaneous fixation of scaphoid fractures with a novel dorsal
technique. Orthop Clin North Am. 2001;32(2):247e261.
r Vol. 41, January 2016