Dow

nloadedfrom

https://journals.lww.com

/jbjsjournalbyyuG

CF+m

L7x/UQ1IYkjsQ

l24BqcohQ32pC

HpY1X0O

xbz6lWklcw

Qn3VbKT9H

y/jlQM43LU

vGDo00ekG

ivFhELEuBjYq+saZbO1+IpR

b+zmoQ

/LAO41BEU

+68mevM

+udNaW

jg4sqx2si8=on

01/03/2019

Downloadedfromhttps://journals.lww.com/jbjsjournalbyyuGCF+mL7x/UQ1IYkjsQl24BqcohQ32pCHpY1X0Oxbz6lWklcwQn3VbKT9Hy/jlQM43LUvGDo00ekGivFhELEuBjYq+saZbO1+IpRb+zmoQ/LAO41BEU+68mevM+udNaWjg4sqx2si8=on01/03/2019

Infantile Supracondylar Humeral Fractures(Patients Less Than Two Years of Age)

Twice as Common in Females and a High Rate of Malunion with LateralColumn-Only Fixation

Charles T. Mehlman, DO, MPH, Jaime Rice Denning, MD, MS, James J. McCarthy, MD, MHCM, and Michael L. Fisher, DO

Investigation performed at Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

Background: The age range for supracondylar humeral fractures spans from 1 to 14 years of age; most publishedstudies have analyzed patients as non-age-segregated cohorts. Some isolated studies focused on the upper agerange, demonstrating a male predominance and more severe fractures. The purpose of the current study was toanalyze a large cohort of patients with surgically treated supracondylar humeral fractures at the low end of the agerange (<2 years of age).

Methods: Patients <2 years of age were identified from surgical billing records. Pin constructs were categorized aslateral column-only fixation or medial and lateral column fixation. All patients were followed through fracture-healing.Substantial loss of reduction was defined as a Baumann angle that changed ‡10� between surgery and healing or as alateral rotation percentage (i.e., Gordon index) of ‡50% at the time of healing. The Fisher exact test was used for statisticalanalysis.

Results: One hundred and three patients met our inclusion criteria. There were 69 female and 34 male patients (a 2:1 female-to-male ratio). Two patients did not have adequate follow-up radiographs. Of the 46 patients with bicolumnarfixation, 5 (11%) demonstrated loss of reduction compared with 20 (36%) of 55 patients with lateral column-only fixation.This difference between the groups was significant (p = 0.005). The group with lateral column-only fixation had 4.7-times-higher odds of loss of reduction (95% confidence interval, 1.6 to 13.8). A subset of patients had in-cast imaging thatallowed calculation of the posterior sagittal cast index (a measure of cast fit). Eight of 15 patients who had a posteriorsagittal cast index of ‡0.20 experienced loss of reduction, while only 1 of 19 patients with a cast index value of <0.20 hadloss of reduction (p = 0.004).

Conclusions: Supracondylar humeral fractures were twice as common in females in this very young cohort. We alsofound a nearly 5-times-higher odds of loss of reduction when fracture fixation was of the lateral column only.

Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Supracondylar humeral fractures are the most commontype of elbow fracture and the most common reasonchildren undergo elbow surgery1-6. Multiple studies have

shown their peak incidence to be among 5 to 6-year-olds, withan age range that reaches from 1 to 14 years7-12. Large studieshave shown that supracondylar fractures occur with nearlyequal frequency among females and males7-9,11,12.

In rare instances, researchers have evaluated age-segregatedsubsets of the supracondylar humeral fracture population. Onestudy focused on pre-teenage patients (defined as >8 years of

age) and identified more severe fracture patterns as well as a 2:1 male-to-female ratio in this older age group13. Another reportshowed a greater than 2.5:1 male-to-female ratio among pa-tients with supracondylar fractures who were 10 to 15 years ofage14. To the best of our knowledge, there are no reported seriesthat have focused on the epidemiology and treatment results atthe low end of the age spectrum. The purpose of the currentstudy was to review a large, single-institution cohort of patientswith supracondylar humeral fractures who were <2 years of ageat the time of injury.

Disclosure: No external funding was received for this study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the onlineversion of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arenaoutside the submitted work (http://links.lww.com/JBJS/E975).

25

COPYRIGHT � 2019 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

J Bone Joint Surg Am. 2019;101:25-34 d http://dx.doi.org/10.2106/JBJS.18.00391

Materials and Methods

This study received institutional review board approval.Patients were identified using 2000 to 2017 billing records

(Current Procedural Terminology [CPT] codes 24538 and24545). Radiographs of patients <2 years of age were reviewed inorder to confirm a supracondylar fracture. Infant was definedaccording to the criteria of the World Health Organization andthe U.S. Food and Drug Administration: 1 month to 2 years ofage15,16. Supracondylar fracture was defined as a fracture thatoccurred in the metaphyseal region of the distal part of thehumerus such that it crossed both medial and lateral columnswhile not extending into the intercondylar region17. Fracturepatterns were classified using the Johns Hopkins supracon-dylar humeral fracture scheme17 (Fig. 1). Fracture displace-ment was categorized according to the modified Gartlandclassification18, and the maximal percentage of translationwas measured on either anteroposterior or lateral views. Theutilization of skeletal surveys was recorded along with asso-ciated findings.

Patients <2 years of age with a confirmed supracondylarfracture were included. We excluded patients with a patho-logic fracture or other known bone-fragility disorder. Demo-graphic, treatment, and complication data were abstractedfrom patient charts. Deviations from the expected postopera-tive treatment course were considered to be complications andwere categorized as minor (not requiring additional surgery) ormajor (requiring additional surgical intervention or long-termmonitoring). Radiographs were analyzed, and the total flu-oroscopic imaging time used for each case was recorded andcategorized as <45 seconds or ‡45 seconds. On the basis ofprior research at our institution, ‡45 seconds was consideredto be a measure of case difficulty19. Pin constructs were cat-egorized as lateral column-only fixation or medial and lateralcolumn fixation, according to the concept illustrated in

Figure 2. Bicolumnar fixation was considered to be achiev-able via 2 lateral-entry pins with appropriate divergence or bytraditional crossed-pin constructs. Although postoperativein-cast radiographs were not routinely obtained, a subsetexisted and were assessed by measuring the ratio of poste-rior cast padding to elbow soft-tissue diameter at the level ofthe antecubital fossa (the posterior sagittal cast index, ameasure of cast fit) and by angular elbow position (whereby a

Fig. 1

Radiographic images demonstrating supracondylar humeral fracture patterns according to the Johns Hopkins fracture classification system17.

Fig. 2

Lateral column and medial column fixation concept applied in the current

study.

26

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

Fig. 3

Images demonstrating the posterior sagittal cast index (the ratio of maximal posterior cast space, i.e., padding, to elbow soft-tissue diameter at the level of

the antecubital fossa) asmeasured on a lateral radiograph (Fig. 3-A) and angular elbow position in a cast (the humeroulnar angle) asmeasured on a lateral

radiograph (Fig. 3-B).

Fig. 4

Figs. 4-A through 4-D Images demonstrating loss of reduction according to the study criteria. Figs. 4-A and 4-B Immediate postoperative anteroposterior

fluoroscopic image (Fig. 4-A) and radiograph from the timeof pin removal showing a change of18� (Fig. 4-B). Figs. 4-Cand4-D Immediatepostoperative lateral

fluoroscopic image showing near-anatomic alignment (Fig. 4-C) and radiograph from the time of pin removal showing a Gordon index of 50% (Fig. 4-D).

27

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

fully extended elbow = 0�) (Fig. 3). The posterior sagittalcast index was categorized as <0.20 or ‡0.20 and the angularelbow position, as <75� or ‡75�. The cast index cutoff wasdetermined on the basis of the subjective clinical judgmentof cast radiographic appearance, and the angular elbowposition cutoff was selected because it represented the lowend of the elbow-position data reported by Mapes andHennrikus20.

Patients were radiographically followed through fracture-healing (a minimum of 1 month of follow-up), with somefollowed longer, according to surgeon preference. TheBaumann angle was measured at the time of surgery and atfracture-healing21, with correction made for the angle of thex-ray beam using the Seattle radioulnar overlap method22.The criteria of Williamson et al. were accepted as the normalrange for the Baumann angle (64� to 81�)23. The primaryoutcome variable was the rate of substantial loss of reduc-tion at the time of fracture-healing. Loss of reduction wasdemonstrated 1 of 2 ways (Fig. 4): by comparing the 2Baumann angle measurements (at the time of surgery andfracture-healing), with substantial loss of reduction definedas a change of ‡10�, with 10� being 3� greater than theestablished measurement error for this angle24; or by a lat-eral rotation percentage (the so-called Gordon index) of‡50% at the time of healing following anatomic or near-anatomic reduction documented at the time of surgery25.The Fisher exact test was used for statistical analysis ofdifferences in proportions. P values of <0.05 were consid-ered significant.

Results

One hundred and three patients met the inclusion criteria,each with a unilateral closed fracture. The average age at

the time of injury was 18 months (range, 9 to 23 months).There were 69 female and 34 male patients (a 2:1 female-to-male ratio). On the basis of the Gartland classification, 37%(38) of the 103 fractures were type II and 63% (65) were typeIII. One patient sustained a type-II flexion injury. No type-IVfractures or open fractures occurred in this series. Preopera-tively, 1 patient was noted to be pulseless (yet perfused) and tohave nerve injury (median nerve), both of which resolvedpostoperatively. No patient in the current series developedcompartment syndrome. A typical fracture pattern was man-ifested by 80% (82) of the patients (Table I), and the mostcommon mechanism of injury was a fall from a household

height (e.g., couch or table) (Table II). Among the 65 type-IIIfractures, 60% (39) demonstrated posteromedial displacement,while the other 40% (26) were posterolateral. The amount ofdisplacement of the type-III fractures averaged 52% (range,2% to 171%). Only 10% (11) of all 103 fractures presentedwith >100% displacement. Lateral-entry pin patterns wereutilized in 93% of the cases, with the remainder involvingcrossed pins.

Two of the 103 patients had inadequate radiographs.The treatment data of the remaining 101 patients showedthat 46 had bicolumnar fixation, 5 (11%) of whom dem-onstrated loss of reduction. Among the 55 patients withfixation of only the lateral column, 20 (36%) demonstratedloss of reduction. This difference between the fixationgroups was significant (p = 0.005). The group with lateralcolumn-only fixation had 4.7-times-higher odds of loss ofreduction (95% confidence interval [CI], 1.6 to 13.8). The 2groups did not differ significantly with respect to the pro-portion of type-III fractures (36 of 55 [66%] for lateralcolumn-only fixation compared with 28 of 46 [61%] forbicolumnar fixation; p = 0.68). Additional characteristics ofthese 2 groups are presented in Table III. Data regarding the25 patients who met the criteria for loss of reduction are

TABLE I Distribution by Johns Hopkins Fracture Pattern17

(N = 103)

Fracture Pattern No.

Typical 82

Medial oblique 12

Lateral oblique 5

High 4

TABLE III Infantile Supracondylar Humeral Fracture PatientCohort Characteristics

Fixation

Lateral ColumnOnly (N = 55)

Lateral 1 MedialColumn (N = 46)

Age at injury* (mo) 18.3 (9-23) 18.8 (12-23)

Sex (%)

Female 61% 74%

Male 39% 26%

Fracture classification (%)

Type III 66% 61%

Type II 34% 39%

Loss of reduction (no. [%]) 20 (36%) 5 (11%)

*The values are given as the mean, with the range in parentheses.

TABLE II Distribution by Mechanism of Injury (N = 103)

Mechanism of Injury No.

Fall from household height 75

Fall from substantial height* 4

Fall at play (FOOSH)† 8

Nonaccidental trauma 4

Unknown 12

*Fall from >10 ft (>3 m). †FOOSH = fall on an outstretched hand.

28

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

presented in Table IV. These patients had postoperativeradiographic follow-up that averaged 3 months (range, 1 to13 months). Thirty-four patients had in-cast radiographs(1 was splinted); 9 of these patients demonstrated loss ofreduction. Images demonstrating the posterior sagittal castindex and angular elbow position for each of these patientsare shown in Figure 5. Eight of 15 patients who had aposterior sagittal cast index of ‡0.20 experienced loss ofreduction, while only 1 of 19 patients with an index value of<0.20 had loss of reduction (p = 0.004). Among the 34patients with in-cast radiographs, the proportion with anangular elbow position of <75� did not differ significantlybetween those with loss of reduction (6 of 9 patients) andthose without loss of reduction (14 of 25 patients) (p = 0.7).We identified a significant difference in recorded fluoro-scopic time (considered to be a measure of case difficulty):48% (12) of the 25 total patients with loss of reduction had

a recorded time of ‡45 seconds compared with 24% (18) ofthe 76 patients who did not experience loss of reduction(p = 0.026).

Skeletal surveys were obtained for 11% (11) of the fullcohort of 103 patients, with an overall rate of 5% (5 of 103)with findings suggestive of nonaccidental trauma. Amongthe patients with radiographic analysis, an 11% rate ofminor complications (11 of 101) was noted (migration of1 or more of the smooth Kirschner wires). Major compli-cations were identified for 3% (3 of 101), each with 1 majorcomplication: 1 patient had deep infection, 1 had medialgrowth arrest, and 1 patient underwent revision fracturesurgery 10 days following the index procedure. No correc-tive osteotomies were performed among the patients in thecurrent series. A total of 8 patients had cubitus varus mea-surements that exceeded 90�, and 4 patients met or exceededa Gordon index of 60%. Additional later radiographs

TABLE IV Patients with Loss of Reduction*

Patient Fracture Fixation Baumann Angle Gordon Index

Age(mo) Sex Side Type

Johns HopkinsClassification

No. ofPins Pattern Column

AtSurgery

AtHealing

AtSurgery

AtHealing

1 18 F Left III (PM) Med oblique 2 Lat entry Lat 78� 91� 0.15 0.17

2 19 F Right III (PL) Typical 2 Lat entry Lat 1 med 79� 75� 0 0.51

3 17 M Right III (PL) Typical 2 Lat entry Lat 75� 74� 0 0.52

4 12 M Left III (PM) Typical 3 Lat entry Lat 1 med 82� 98� 0 0

5 16 F Right II Typical 2 Lat entry Lat 73� 91� 0 0.16

6 9 M Right III (PM) Med oblique 3 Lat entry Lat 80� 94� 0 0

7 15 M Left II (flex) High 2 Lat entry Lat 80� 63� 0 0

8 20 M Left III (PM) Med oblique 2 Lat entry Lat 77� 89� 0 0.22

9 16 F Right III (PL) Typical 2 Lat entry Lat 76� 78� 0.06 0.61

10 22 F Right III (PL) Typical 2 Lat entry Lat 72� 93� 0 0.30

11 23 F Right III (PM) Typical 3 Lat entry Lat 1 med 73� 72� 0.10 0.50

12 23 F Right II Lat oblique 3 Lat entry Lat 1 med 82� 72� 0 0

13 21 M Left III (PM) Typical 2 Lat entry Lat 76� 63� 0 0.14

14 14 F Left III (PL) Typical 2 Lat entry Lat 78� 92� 0 0.22

15 21 M Left III (PL) Typical 3 Lat entry Lat 74� 86� 0.09 0.47

16 12 F Right III (PL) Lat oblique 2 Lat entry Lat 75� 88� 0.14 0.21

17 18 F Right III (PL) Typical 3 Lat entry Lat 67� 76� 0.18 0.62

18 20 F Right III (PM) Typical 2 Lat entry Lat 83� 94� 0 0

19 18 F Left III (PL) Typical 3 Lat entry Lat 75� 87� 0 0.58

20 18 F Left III (PL) Typical 2 Lat entry Lat 64� 80� 0.08 0.10

21 17 M Right II Typical 3 Lat entry Lat 1 med 78� 59� 0 0.09

22 19 F Left III (PL) Typical 2 Lat entry Lat 77� 74� 0 0.82†

23 18 F Left III (PM) Typical 2 Lat entry Lat 78� 91� 0.14 0.12

24 22 M Left III (PL) Typical 3 Lat entry Lat 75� 76� 0 0.60

25 23 F Right III (PL) Typical 2 Lat entry Lat 75� 86� 0 0.19

*PM = posteromedial, Med = medial, Lat = lateral, PL = posterolateral, and flex = flexion. †Loss of reduction detected at 1 week of follow-up; thepatient underwent revision surgery.

29

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

revealed that 10% (10) of the 101 patients sustained a sec-ond elbow fracture: 3 experienced a second ipsilateralsupracondylar fracture, 3 had an ipsilateral lateral condylarfracture, and 1 patient experienced ipsilateral simultaneousfractures of the olecranon and radial neck. Two otherpatients had later contralateral lateral condylar fractures,

and 1 patient had a later contralateral supracondylar humeralfracture.

Discussion

We identified 1 previous report that proposed to focuson infantile supracondylar humeral fractures, a 1967

Fig. 5

Figs. 5-A through5-INine patients from Table IVwho had postoperative “in-cast” imaging andwent on to experience loss of reduction. Fig. 5-A Patient 6.

Fig. 5-B Patient 7. Fig. 5-C Patient 15. Fig. 5-D Patient 17. Fig. 5-E Patient 19. Fig. 5-F Patient 21. Fig. 5-G Patient 22. Fig. 5-H Patient 24.

Fig. 5-I Patient 25.

30

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

paper by Macafee entitled “Infantile Supracondylar Frac-ture.”26 All 3 of the included patients (a 19-month-old girl,a 12-month-old boy, and a 2-day-old boy) actually haddisplaced transphyseal fractures, and Macafee was simplyusing the typical, albeit confusing, terminology of the day26.However, this remains pertinent as it illustrates the need to bealert for the diagnosis of transphyseal fracture in this agegroup. In fact, orthopaedic surgeons should doubt the diag-nosis of supracondylar fracture in such young children untilthey have been able to personally review injury radiographs.All of the patients included in the current study had truesupracondylar humeral fractures, with fracture lines trans-gressing the metaphyseal portion of the distal part of thehumerus17.

Also important in the treatment of such very young frac-ture patients is to remain cognizant of the possibility of childabuse. Findings consistent with child abuse were confirmed in5% of the patients in the current study. This is higher than the1.4% rate previously reported among 69 patients with dis-placed supracondylar humeral fractures who were £4 years ofage27. A large U.S. national database study demonstrated thatmost children with fractures due to nonaccidental trauma are<2 years of age28. It is interesting to note that, among a cohort of115 fracture patients of <2 years of age, there was a 17% rate ofchild abuse and no abused child had a supracondylar humeralfracture29.

We identified a 2:1 female-to-male ratio in the veryyoung cohort of the current study. A similar female predomi-nance in the youngest supracondylar humeral fracture agegroup was documented by Farnsworth and colleagues. Theirreport included 40 patients in the age group of 0 to 1 year, 29girls and 11 boys (a 2.6:1 female-to-male ratio)30. They alsostated that their supracondylar fracture patients who were £3years of age tended to fall from 3 to 6-foot (approximately 1 to2-meter) heights from household objects30. This mechanismalso predominated in the present study, as 73% of the patientsfell from such relatively low household heights.

Dissimilar findings among such young patients havebeen reported by other authors. In a series of all long-bonefractures (not just supracondylar fractures) in 103 patients <3years of age, no substantially increased frequency for maleversus female patients was noted (53% male, 47% female)31.The most common fractures in the same series were radial andulnar fractures (35%); supracondylar humeral fractures ac-counted for only 14%31. Roda et al. reported on an even youngerseries (all <12 months of age) of all fracture patients presentingto a pediatric emergency department, noting no dominance bysex. They found that 57% (58 of 101) were male and that skullfractures were most common (57%); only 1 supracondylarhumeral fracture (an incomplete fracture pattern) was noted inthe series32. The current study differs because of its homoge-nous group of infantile patients with supracondylar humeralfractures.

As mentioned earlier, a previous study of supracondylarfractures, by Fletcher et al., focused on an older age group andreported a 2:1 male-to-female ratio13. Reporting on 118 ado-

lescents (10 to 17 years of age) with a variety of complexsupracondylar fractures, Bell et al. reported a 3.5:1 male-to-female ratio14. It is easy to rationalize these higher supracon-dylar fracture injury rates for males by considering the moreviolent sporting activities that such boys may participate in. Inour study, it is unclear why a strong female preponderance ofthese fractures would occur, in an age group for which onewould expect the same risk of injury. A possible explanationmay relate to relative bone-density differences demonstrated byvery young boys and girls as well as rates of elbow hyperex-tension among the 2 sexes. It is well known that infancy is anintensely dynamic time period regarding bone physiology, thehallmark being complete skeletal turnover33,34. It has also beenshown that, on average, females in this very young age rangehave significantly lower bonemineral content than domales35.Additionally, there is the long-held contention that ligamen-tous laxity (with associated elbow hyperextension) is anindependent risk factor for extension-type supracondylarfractures36, and the rate of such ligamentous laxity has beenshown to be greater among preschool females37. These factorscould lead one to hypothesize that this variable bone mineralcontent combined with the baseline increased biomechanicalsusceptibility to elbow hyperextension may be an explanationfor the higher fracture rate that we identified among femalepatients.

We discovered rates of loss of reduction at the time ofhealing that we found concerning: 36% in the group withlateral column-only fixation and still an 11% rate in thegroup with bicolumnar fixation. Explanations for thesefindings are certainly multifactorial. Clinical experience hasshown that treating such young patients with supracondylarfractures presents technical challenges because of the patients’smaller size.We believe that the differences in total fluoroscopictime documented in the current study support this contention.The target is smaller, and the potential bone-weakening effectof multiple pin passes may be magnified. In addition, althoughfar from definitive, our data do suggest that there is a greaterprice to pay when postoperative immobilization of theseinfants is less than form-fitting. In fracture terminology, thepins are the primary fracture-stabilization tool and the cast iseffectively a neutralization device.

Loss of reduction of surgically treated supracondylarfractures has been studied by others. In 1 large cohort ofnon-age-segregated patients, a 4.2% (8 of 192) rate of loss ofreduction (defined as a >10� change in non-rotation-correctedBaumann angle) was documented38. Pin spread (effectively ameasure of unicolumnar versus bicolumnar fixation) wasthe only predictor of loss of reduction, while no age-relatedcorrelation was found38. Primarily using the criteria of repeatsurgical intervention, with no specific radiographic defini-tion being offered, Sankar et al. reported a 2.9% (8 of 271)rate of loss of reduction39. They attributed all cases of lossof reduction to what they considered technical errors in pinfixation. Similar to the previously cited series38,39, we noted thatno patient with loss of reduction in the current study haddeformity of sufficient magnitude to trigger corrective osteotomy.

31

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

In fact, our patients’ very young age certainly enhanced theirremodeling potential (Figs. 6 and 7), and we would thus dis-courage early osteotomy.

Our findings must be interpreted within the context ofthe study design. This was a retrospective study with theprimary outcome variable being radiographic loss of

Fig. 7

Figs. 7-A, 7-B, and 7-C Short-term successful remodeling of the Gordon index (rotational spike) in Patient 17 of Table IV. Fig. 7-A Fluoroscopic image

showingnear-anatomic reduction.Fig. 7-BRadiograph from4weeksof follow-up demonstrating a large rotational spike.Fig. 7-CRadiograph from8months

of follow-up showing adequate remodeling.

Fig. 6

Figs. 6-A, 6-B, and6-CShort-termsuccessful remodeling of theBaumannangle in Patient 18 of Table IV. Fig. 6-A Fluoroscopic imagewith persistentmedial

column impaction. Fig. 6-B Radiograph at 4 weeks of follow-up demonstrating an 11� increase in the Baumann angle and clear cubitus varus.

Fig. 6-C Radiograph at 8 months of follow-up showing the re-establishment of a valgus Baumann angle.

32

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

reduction. Our study group was just over 100 patients, so alarger cohort might have revealed different male-to-femaleproportions or different radiographic outcomes. We alsorecognize that our subgroup analysis consisted of only 34patients with in-cast radiographs, thus placing it at risk ofstatistical instability. It is possible that functional outcomemeasures might have produced different findings. However,to the best of our knowledge, this study represents the largestreported series of infantile patients with supracondylar humeralfractures, and we believe that our findings offer valid epidemi-ologic and treatment-related information. Longer follow-upfor all patients with malunion would also have been ideal inorder to fully assess remodeling potential.

In summary, we found that infantile supracondylarhumeral fractures were twice as common in female versus malepatients. Most patients sustained the fractures from falls fromhousehold heights and demonstrated typical fracture patternswith modest displacement (only 10% were >100% displaced).Nearly 5-times-higher odds of loss of reduction were docu-mented for lateral column-only fixation compared with medialand lateral column fixation. A subgroup analysis also indicated

that the surgical care of these infantile patients is technicallychallenging, and that every effort should be made to utilizeform-fitting postoperative immobilization. n

Charles T. Mehlman, DO, MPH1

Jaime Rice Denning, MD, MS1

James J. McCarthy, MD, MHCM1

Michael L. Fisher, DO2

1Cincinnati Children’s Hospital Medical Center, University of CincinnatiCollege of Medicine, Cincinnati, Ohio

2Summa Health System/Western Reserve Hospital, Ohio UniversityHeritage College of Osteopathic Medicine, Cuyahoga Falls, Ohio

E-mail address for C.T. Mehlman: Charles.mehlman@cchmc.org

ORCID iD for C.T. Mehlman: 0000-0001-8292-2508ORCID iD for J.R. Denning: 0000-0003-1743-4801ORCID iD for J.J. McCarthy: 0000-0001-9796-7207ORCID iD for M.L. Fisher: 0000-0002-4631-6874

References

1. Cheng JC, Shen WY. Limb fracture pattern in different pediatric age groups: astudy of 3,350 children. J Orthop Trauma. 1993;7(1):15-22.2. Emery KH, Zingula SN, Anton CG, Salisbury SR, Tamai J. Pediatric elbow frac-tures: a new angle on an old topic. Pediatr Radiol. 2016 Jan;46(1):61-6. Epub 2015Jul 28.3. Holt JB, Glass NA, Bedard NA, Weinstein SL, Shah AS. Emerging U.S. nationaltrends in the treatment of pediatric supracondylar humeral fractures. J Bone JointSurg Am. 2017 Apr 19;99(8):681-7.4. Houshian S, Mehdi B, Larsen MS. The epidemiology of elbow fracture in children:analysis of 355 fractures, with special reference to supracondylar humerus frac-tures. J Orthop Sci. 2001;6(4):312-5.5. Sinikumpu JJ, Pokka T, Sirvio M, Serlo W. Gartland Type II supracondylar humerusfractures, their operative treatment and lateral pinning are increasing: a population-based epidemiologic study of extension-type supracondylar humerus fractures inchildren. Eur J Pediatr Surg. 2017 Oct;27(5):455-61. Epub 2016 Dec 9.6. Vallila N, Sommarhem A, Paavola M, Nietosvaara Y. Pediatric distal humeralfractures and complications of treatment in Finland: a review of compensationclaims from 1990 through 2010. J Bone Joint Surg Am. 2015 Mar 18;97(6):494-9.7. Barr LV. Paediatric supracondylar humeral fractures: epidemiology, mechanismsand incidence during school holidays. J Child Orthop. 2014 Mar;8(2):167-70. Epub2014 Mar 19.8. Edmonds EW, Roocroft JH, Mubarak SJ. Treatment of displaced pediatricsupracondylar humerus fracture patterns requiring medial fixation: a reliable andsafer cross-pinning technique. J Pediatr Orthop. 2012 Jun;32(4):346-51.9. Khoshbin A, Leroux T, Wasserstein D, Wolfstadt J, Law PW, Mahomed N, WrightJG. The epidemiology of paediatric supracondylar fracture fixation: a population-based study. Injury. 2014 Apr;45(4):701-8. Epub 2013 Oct 17.10. Mehlman CT, Crawford AH, McMillion TL, Roy DR. Operative treatment ofsupracondylar fractures of the humerus in children: the Cincinnati experience. ActaOrthop Belg. 1996;62(Suppl 1):41-50.11. Mehlman CT, Strub WM, Roy DR, Wall EJ, Crawford AH. The effect of surgicaltiming on the perioperative complications of treatment of supracondylar humeralfractures in children. J Bone Joint Surg Am. 2001 Mar;83(3):323-7.12. Sinikumpu JJ, Victorzon S, Pokka T, Lindholm EL, Peljo T, Serlo W. The long-termoutcome of childhood supracondylar humeral fractures: a population-based follow upstudy with aminimum follow up of ten years and normal matched comparisons. BoneJoint J. 2016 Oct;98-B(10):1410-7.13. Fletcher ND, Schiller JR, Garg S, Weller A, Larson AN, KwonM, Browne R, CopleyL, Ho C. Increased severity of type III supracondylar humerus fractures in the preteenpopulation. J Pediatr Orthop. 2012 Sep;32(6):567-72.14. Bell P, Scannell BP, Loeffler BJ, Brighton BK, Gaston RG, Casey V, Peters ME,Frick S, Cannada L, Vanderhave KL. Adolescent distal humerus fractures: ORIFversus CRPP. J Pediatr Orthop. 2017 Dec;37(8):511-20.

15. Knoppert D, Reed M, Benavides S, Totton J, Hoff D, Moffett B, Norris K, Vaillan-court R, Aucoin R, Worthington M. Position paper: pediatric age categories to be usedin differentiating between listing on a model essential medicines list for children.http://archives.who.int/eml/expcom/children/Items/PositionPaperAgeGroups.pdf. Accessed 2017 Oct 8.16. U.S. Food and Drug Administration. Pediatric exclusivity study age group. https://www.fda.gov/Drugs/DevelopmentApprovalProcess/FormsSubmissionRequirements/ElectronicSubmissions/DataStandardsManualmonographs/ucm071754.htm. Ac-cessed 2017 Oct 8.17. Bahk MS, Srikumaran U, Ain MC, Erkula G, Leet AI, Sargent MC, Sponseller PD.Patterns of pediatric supracondylar humerus fractures. J Pediatr Orthop. 2008 Jul-Aug;28(5):493-9.18. Mencio GA. Fractures and dislocations about the elbow. In: Menci G, Swiont-kowski M, editors. Green’s skeletal trauma in children. 5th ed. Philadelphia:Elsevier/Saunders; 2015. p 187-8.19. Eismann EA, Wall EJ, Thomas EC, Little MA. Direct beam radiation exposureto surgeons during pinning of supracondylar humerus fractures: does C-armposition and the attending surgeon matter? J Pediatr Orthop. 2014 Mar;34(2):166-71.20. Mapes RC, Hennrikus WL. The effect of elbow position on the radial pulsemeasured by Doppler ultrasonography after surgical treatment of supracondylarelbow fractures in children. J Pediatr Orthop. 1998 Jul-Aug;18(4):441-4.21. Camp J, Ishizue K, Gomez M, Gelberman R, Akeson W. Alteration of Baumann’sangle by humeral position: implications for treatment of supracondylar humerusfractures. J Pediatr Orthop. 1993 Jul-Aug;13(4):521-5.22. Pace JL, Wiater B, Schmale G, Jinguji T, Bompadre V, Krengel W 3rd. Baumannangle and radial-ulnar overlap: a radiographic study to control for the angle of the x-ray beam. J Pediatr Orthop. 2012 Jul-Aug;32(5):467-72.23. Williamson DM, Coates CJ, Miller RK, Cole WG. Normal characteristics of theBaumann (humerocapitellar) angle: an aid in assessment of supracondylar frac-tures. J Pediatr Orthop. 1992 Sep-Oct;12(5):636-9.24. Silva M, Pandarinath R, Farng E, Park S, Caneda C, Fong YJ, Penman A. Inter- andintra-observer reliability of the Baumann angle of the humerus in children with supra-condylar humeral fractures. Int Orthop. 2010 Apr;34(4):553-7. Epub 2009 May 8.25. Gordon JE, Patton CM, Luhmann SJ, Bassett GS, Schoenecker PL. Fracturestability after pinning of displaced supracondylar distal humerus fractures in chil-dren. J Pediatr Orthop. 2001 May-Jun;21(3):313-8.26. Macafee AL. Infantile supracondylar fracture. J Bone Joint Surg Br. 1967 Nov;49(4):768-70.27. Pandya NK, Baldwin KD, Wolfgruber H, Drummond DS, Hosalkar HS. Humerusfractures in the pediatric population: an algorithm to identify abuse. J Pediatr OrthopB. 2010 Nov;19(6):535-41.

33

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)

28. Loder RT, Feinberg JR. Orthopaedic injuries in children with nonaccidentaltrauma: demographics and incidence from the 2000 Kids’ Inpatient Database. JPediatr Orthop. 2007 Jun;27(4):421-6.29. Leaman LA, Hennrikus WL, Bresnahan JJ. Identifying non-accidental frac-tures in children aged <2 years. J Child Orthop. 2016 Aug;10(4):335-41. Epub2016 Jun 23.30. Farnsworth CL, Silva PD, Mubarak SJ. Etiology of supracondylar humerus frac-tures. J Pediatr Orthop. 1998 Jan-Feb;18(1):38-42.31. Taitz J, Moran K, O’Meara M. Long bone fractures in children under 3 years ofage: is abuse being missed in emergency department presentations? J PaediatrChild Health. 2004 Apr;40(4):170-4.32. Roda D, Trenchs V, Curcoy AI, Martınez AD, Pou J, Luaces C. Epidemiology offractures in children younger than 12 months. Pediatr Emerg Care. 2017 May 1.Epub 2017 May 1.33. Bollen AM, Eyre DR. Bone resorption rates in children monitored by theurinary assay of collagen type I cross-linked peptides. Bone. 1994 Jan-Feb;15(1):31-4.

34. Buckwalter JA, Glimcher MJ, Cooper RR, Recker R. Bone biology. II: formation, form,modeling, remodeling, and regulation of cell function. Instr Course Lect. 1996;45:387-99.35. Kalkwarf HJ, Zemel BS, Yolton K, Heubi JE. Bone mineral content and density ofthe lumbar spine of infants and toddlers: influence of age, sex, race, growth, andhuman milk feeding. J Bone Miner Res. 2013 Jan;28(1):206-12.36. Nork SE, HennrikusWL, Loncarich DP, GillinghamBL, Lapinsky AS. Relationshipbetween ligamentous laxity and the site of upper extremity fractures in children:extension supracondylar fracture versus distal forearm fracture. J Pediatr Orthop B.1999 Apr;8(2):90-2.37. Lamari NM, Chueire AG, Cordeiro JA. Analysis of joint mobility patterns amongpreschool children. Sao Paulo Med J. 2005 May 2;123(3):119-23. Epub 2005 Jul 8.38. Pennock AT, Charles M, Moor M, Bastrom TP, Newton PO. Potential causes ofloss of reduction in supracondylar humerus fractures. J Pediatr Orthop. 2014 Oct-Nov;34(7):691-7.39. Sankar WN, Hebela NM, Skaggs DL, Flynn JM. Loss of pin fixation in displacedsupracondylar humeral fractures in children: causes and prevention. J Bone JointSurg Am. 2007 Apr;89(4):713-7.

34

THE JOURNAL OF BONE & JOINT SURGERY d J B J S .ORG

VOLUME 101-A d NUMBER 1 d JANUARY 2, 2019INFANT ILE SUPRACONDYLAR HUMERAL FRACTURES (PAT IENTS LES S

THAN TWO YEARS OF AGE)