Musculoskeletal Injuries in National Hockey League Goaltenders

Abstract

Background:

Hockey goaltenders are exposed to unique in-game physical demands, different from those experienced by other positional players. While previous studies have identified the injuries experienced by elite goaltenders, these have been limited by the use of secondary data or data from small samples over relatively short time intervals.

Purpose:

To quantify the type and number of injuries occurring in National Hockey League (NHL) goaltenders and the time lost due to these injuries.

Study Design:

Descriptive epidemiology study.

Methods:

Data from the NHL’s Athlete Health Management System was analyzed between the 2009-2010 and 2022-2023 seasons. The total number of each musculoskeletal injury, the total days missed (TDM) (a surrogate measure for injury severity), and the exposure-adjusted injury rate (EAIR) was quantified. For each individual diagnosis, the total number of injuries, mean, median, standard deviation, and minimum and maximum (range) TDM were calculated.

Results:

There were 1585 musculoskeletal injuries recorded over 14 NHL seasons resulting in 31,406 TDM. The number of total injuries per year was stable across the data collection period. The EAIR was higher in the preseason (69.4) and postseason (46.8) compared with the regular season (17.1). The majority of the injuries occurred during games (n = 904), followed by practice and off-ice activities (n = 681). The most common mechanisms were noncontact (n = 673), incidental contact (n = 252), and “unknown/other” (n = 374). Proximal adductor injuries accounted for the largest number of injuries (n = 371), leading to the highest total TDM (n = 6126). Concussions (n = 110) and MCL sprains (n = 95) were the second and third most common injuries, respectively.

Conclusion:

The most common injuries to NHL goaltenders are proximal adductor injuries, concussions, and MCL sprains. Attention should be paid to the significant number of injuries occurring in the preseason and practice periods.

Ice hockey goaltenders face unique physical demands, unlike those experienced by positional players. These include biomechanically demanding movements that require frequent drops to the knees, rapid lateral movements, and extreme hip ranges of motion. ²⁵ Bell et al ⁴ identified 17 movements unique to National Hockey League (NHL) goaltenders and determined that these are performed >200 times per game. Goaltender-specific movements have been associated with a range of lower extremity pathologies including intra-articular hip injury, adductor muscle strain, athletic pubalgia, and femoroacetabular impingement syndrome. ⁵ In elite goaltenders, the cumulative incidence for all hip and groin pathologies over the course of a single season has been reported to be up to 70%, with 36% incurring hip and groin pathologies that resulted in a moderate or severe reduction in training volume or the inability to participate in games and practices. ⁴¹

While previous studies have identified some of the injuries experienced by professional goaltenders, these have been limited by their use of secondary data or small sample sizes over relatively short time intervals (eg, 1-2 seasons).^26,43 The lack of a comprehensive data set quantifying the type, severity, and epidemiology of these injuries over multiple seasons makes it difficult for sports medicine professionals to effectively identify and reduce injury risk in this population and manage return to play.

Therefore, the purpose of this study was to quantify the (1) number and incidence of the most commonly occurring musculoskeletal injuries in NHL goaltenders and (2) time lost due to these injuries.

Methods

Data Acquisition

A retrospective analysis was performed on 14 years of consecutive injuries to NHL goaltenders sustained between the 2009-2010 and 2022-2023 seasons. The NHL consists of 32 teams, each with a roster including at least 2 goaltenders. During the assessment period, 2 expansion teams were added to the league (Las Vegas Golden Knights [2017] and Seattle Kraken [2021]).

Injury data were obtained from the Athlete Health Management System (AHMS) after approval by the NHL. The AHMS is a centralized database that contains the medical information from each team’s electronic medical records. After an injury, team medical personnel categorize the events of the injury by selecting the appropriate identifier from a set of standardized descriptions and may additionally input free-text information. The data set provided for this study included 22 different categories describing the injury and its mechanism (Table 1). All data were anonymized for protection of player privacy.

Table 1.

Summary of the Data Categories Provided From the AHMS ^a

Category	Examples of Subcategories/Descriptors
Season b	2009-2010 to 2022-2023
Section b	Off-season, preseason, regular season, postseason
Session b	Practice, off-ice, non–hockey related, game, scrimmage
Regular season MGL reported date	Regular season games missed within a single season
Season MGL reported date	Regular season and playoff games missed within a single season
Total MGL reported date	Total games missed including regular season and playoffs
MGL (reported end date)	MGL based on the return-to-play date
Days reported date b	Total days missed regardless of point in season
Days reported date– season	Total regular season days missed and can span multiple seasons
Event type b	Injury acute, injury chronic
Side	Right, left, both, NA
Diagnosis b	Proximal adductor strain, nonspecific low back pain, medial rotator sprain, ankle contusion, etc
Region b	Head, neck, shoulder, upper arm, elbow, forearm, etc
Structure	Muscle/tendon, nonspecific, bone, superficial tissue/skin, ligament/joint capsule, nervous, cartilage/synovium/bursae
Pathology type	Strain, fracture–acute, contusion, swelling, 1st-degree sprain–acute, etc
SMDCS code	Sport Medicine Diagnostic Coding System
ICD-9 code	International Classification of Diseases, 9th Revision codes
Exacerbated	Player had ongoing injury that was exacerbated
Reinjury b	Player recovered from similar injury and now reinjured
Period b	1st, 2nd, 3rd (only recorded if injury occurred during a game)
Category b	Hit by puck, unknown, incidental contact with another player or their equipment, noncontact, etc
Category detail	Shot, boards, high sticking, pass, slashing

^aAHMS, Athlete Health Management System; MGL, man-games lost; NA, not applicable.

^bCategories that were included in the current analysis.

While several variables may quantify time lost due to injury, for this study, total days missed (TDM) was used to represent the severity of the injury and the effect on goaltender performance and health. TDM accounts for all time away from hockey-related activity (eg, pre- and postseason activities, practices, and days in between) and is not limited to the number of in-season games during which a goaltender was inactive (man-games lost).

Data Assessment

The data were reviewed by 2 authors (S.K. and E.C.). Free-text data from team medical reports were assessed and grouped into singular diagnoses. Inclusion criteria were injuries involving the musculoskeletal system and injuries that resulted in ≥1 TDM on average. Medical illnesses (eg, cardiovascular, neurological, and infectious disease) and injuries that were nonmusculoskeletal in nature were excluded. Concussions and transient brachial plexus injuries (eg, “stingers”) were separated from other neurological disorders (eg, headaches and migraines) and included in the analysis due to their marked relevance and frequency of occurrence in goaltenders. Contusions and injuries that did not have a specific diagnosis but described regional anatomy (eg, “other ankle injury” and “other knee injury”) were excluded from the main cohort and analyzed separately.

Diagnoses that were similar in nature but input as separate entries were combined (eg, lateral ankle ligament sprain and anterior talofibular ligament sprain). Similarly, injuries on a pathological continuum, such as proximal adductor spasm, strain, and tear/rupture, were grouped into 1 overarching diagnosis (proximal adductor injury) and subanalyzed individually. Additionally, the session (eg, practice, game, and off-ice) and period categories were combined and analyzed together as they both describe when the injury occurred relative to gameplay. Within these categories, injuries that were classified as “unknown” or “non–hockey related” were further combined. Finally, with respect to the injury category (ie, mechanism), the “other” and “unknown” and the “gym/weight room,” “workout area,” and “locker room” classifications were grouped together. Intentional contact includes “fighting” and “body checking.” The grouping process was completed by an orthopaedic surgery resident (S.K.) and approved by a senior, fellowship-trained orthopaedic surgeon and NHL team physician (J.T.) before aggregation.

Statistical Analysis

Descriptive statistics were performed summarizing the number of injuries and the total days lost to injury. After the initial analysis, a z score assessment was conducted on the TDM, for each diagnosis separately, to identify the presence of statistical outliers (z score >±1.96 was considered an outlier). As data were input by third parties, it was not possible to determine the reason a data point may be presenting as an outlier, and therefore the decision was made to present the data with and without the statistical outliers. Each individual injury type, as well as the total number of injuries, was plotted over the 14 years. A simple linear regression was performed on these relationships to quantify the rate of change in the number of injuries (represented by the slope).

The exposure-adjusted injury rates (EAIRs) per 1000 athlete-exposures were also quantified:

$$\mathrm{EAIR}=\left(\frac{\mathrm{number}\mathrm{of}\mathrm{injuries}}{\mathrm{exposure}}\right)\times 1000$$

Exposure was calculated by first quantifying the total number of games played over the 14-year assessment period, including both regular season and playoff games. This number was then multiplied by 2 to account for the majority of situations in which there are 2 goalies on the ice at any given time. In addition to the total exposure, separate EAIRs were also calculated for each section (preseason, regular season, and postseason) of the season, and the exposure was calculated separately for those times.

Further analysis was performed on the top 20 diagnoses. This cutoff was chosen as these represented >75% of all TDM due to injury. Extended analyses included the timing of injury within gameplay, mechanism of injury (eg, noncontact and hit by puck), section of the season (preseason, regular season, and postseason), and injury trends over time.

All data analyses were completed using IBM SPSS statistical analysis software (Version 29; IBM Corp).

Results

The TDM data presented throughout this section are those with the outliers included, as there was not enough information available to justify the full exclusion of a statistical outlier.

Over 14 NHL seasons, 1585 injuries across 112 individual diagnoses met the inclusion criteria, resulting in 31,406 TDM. The mean number of injuries per year was 113.3 (SD, 16.6), and the overall injury rate was 22.6 injuries per 1000 athlete-exposures.

Proximal adductor injuries were the most common (n = 371), representing 23.4% of all goaltender injuries. This was followed by concussions (n = 110; 6.9%), MCL injuries (n = 95; 6.0%), cervical muscle strain (n = 93; 5.9%), and low back injury (n = 84; 5.3%). A complete list can be found in Table 2.

Table 2.

Summary of the Top 20 Injuries Sustained by NHL Goaltenders Over the Most Recent 14 NHL Seasons ^a

Diagnosis	No. of Injuries	% of Total Injuries	TDM	Mean, days	Range, days	Median, days	SD, days	% Reinjury	Mean With Outliers Removed, days	No. of Cases Omitted, n (%) b
Proximal adductor injury	371	23.4	6126	16.5	0-193	8	28.3	3.4	10.9	20 (5.4)
Spasm	27	1.7	66	2.4	0-18	0	5.1	0	0.8	3 (11.1)
Tendinopathy	5	0.3	41	8.2	0-17	12	7.8	0	8.2	00 (0)
Strain	328	20.7	5302	16.2	0-193	8	27.2	3.4	11.5	14 (4.3)
Tear	11	0.7	717	65.2	0-136	88	45.9	0	65.2	00 (0)
Concussion	110	6.9	2746	25.0	1-620	12	63.0	0	16.8	3 (2.7)
MCL sprain	95	6.0	3152	63.9	0-257	18	44.3	0	23.5	7 (7.4)
1st degree	70	4.4	1306	18.7	0-130	7	27.2	0	13.4	4 (5.7)
2nd degree	23	1.4	1643	71.4	12-257	53	56	0	63	1 (4.3)
3rd degree	2	0.1	203	101.5	45-158	101.5	79.9	0	101.5	00 (0)
Cervical strain/whiplash	93	5.9	346	3.7	0-189	0	19.7	0	1.7	1 (1.1)
Low back pain	84	5.3	677	8.1	0-190	0	29.7	8.3	3.7	2 (2.4)
Intra-articular hip injury	73	4.6	1809	24.8	0-240	0	55.6	25.2	10.4	6 (8.2)
Hip impingement syndrome	47	3.0	835	17.8	0-196	0	47.9	2.1	4.1	4 (8.5)
Hip labral tear	26	1.6	974	37.5	0-240	0.5	66.4	23.1	23.3	2 (7.7)
Hamstring injury	47	3.0	921	19.6	0-165	1	41.6	0	6.3	5 (10.6)
Spasm	4	0.2	8	2	0-4	2	2.3	0	2	00 (0)
Tendinopathy (biceps)	4	0.2	67	16.8	0-66	0.5	32.8	0	16.8	00 (0)
Strain	38	2.4	818	21.5	0-165	1	45	0	10.8	3 (7.9)
Tear/rupture	1	0.1	28	28	28-28	NA	NA	0	28	00 (0)
Meniscal tears	39	2.4	2912	74.7	0-1009	28	166.3	5.6	50.1	1 (2.6)
Medial	21	1.3	691	32.9	0-198	24	43.9	0	24.7	1 (4.8)
Lateral	18	1.1	2221	123.4	0-1009	0	234.3	5.6	71.3	1 (5.6)
Lacerations	39	2.4	64	1.64	0-32	0	5.5	0	1.64	00 (0)
Face	26	1.6	2	0.08	0-2	0	0.392	0	0.08	00 (0)
Hand	5	0.3	17	3.4	0-7	3	3.5	0	3.4	00 (0)
Thigh	3	0.2	32	10.7	0-32	0	18.5	0	10.7	00 (0)
Leg	3	0.2	11	3.7	0-11	0	6.3		3.7	00 (0)
Ankle	2	0.1	2	1	0-2	1	1.4	0	1	00 (0)
Rotator cuff injury	38	2.4	508	13.4	0-260	0	53.2	0	1.1	2 (5.3)
Strain	36	2.3	281	7.8	0-260	0	43.3	0	0.6	1 (2.8)
Tear	2	0.1	227	113.5	0-208	113.5	133.6	0	19	00 (0)
Lateral ligament sprains	32	2.0	437	13.7	0-114	3	25.8	5.9	8.3	2 (6.3)
CFL 1st degree	17	1.1	89	5.2	0-60	1	14.3	5.9	1.8	1 (5.9)
CFL 2nd degree	3	0.2	144	48	11-114	19	57.3	0	48	00 (0)
ATFL	12	0.8	204	17	0-75	8	23.1	0	11.7	1 (8.3)
Abdominal wall injuries	32	2.0	1296	51.8	0-206	8	68.6	0	38.8	2 (8)
Abdominal wall strain	14	0.9	240	17.1	0-116	0	32.6	0	9.5	1 (7.1)
Rectus abdominus strain	7	0.4	41	5.9	0-5	1	10.7	0	5.9	00 (0)
Sports hernia	11	0.7	1156	96	0-206	111	78	0	96	00 (0)
Finger fractures	28	1.8	205	7.3	0-48	0.5	11.5	0	5.8	1 (3.6)
Distal phalanx	26	1.6	187	7.2	0-48	0.5	11.7	0	5.6	1 (3.8)
Middle phalanx	2	0.1	18	9	0-18	9	12.7	0	9	00 (0)
High ankle sprain	27	1.7	2141	79.3	0-333	42	94.6	5.3	53	3 (11.1)
1st degree	19	1.2	1212	63.8	0-333	28	86.5	5.3	48.8	1 (5.3)
2nd degree	8	0.5	929	116.1	0-272	62	108.8	0	116.1	00 (0)
Paralumbar muscle strain	27	1.7	55	2	0-16	1	3.7	0	1.1	2 (7.4)
Patellar tendinopathy	25	1.6	44	1.8	0-18	0	4.4	0	0.6	2 (8.0)
Iliopsoas strain	24	1.5	55	2.3	0-25	0	5.4	0	1.3	1 (4.2)
Wrist sprain	22	1.4	187	8.5	0-178	0	37.8	0	0.43	1 (4.5)
Elbow UCL sprain	20	1.3	59	2.3	0-39	0	9	0	1.1	1 (5.0)
1st degree	19	1.2	45	2.4	0-39	0	8.9	0	0.3	1 (5.3)
2nd degree	1	0.1	14	14	14-14	NA	NA	0	14	00 (0)
Quadriceps strain	18	1.1	128	7.1	0-38	1	10.7	0	5.3	1 (5.6)

^aATFL, anterior talofibular ligament; CFL, calcaneal fibular ligament; MCL, medial collateral ligament; NHL, National Hockey League; TDM, total days missed; UCL, ulnar collateral ligament.

^bIndividual cases with z scores >±1.96 (threshold to be considered outliers) were omitted on a per-diagnosis basis to calculate the mean with no outliers, as stated in the Methods section.

The individual injuries that resulted in the greatest mean TDM among goaltenders were lateral meniscal tears (123.4 days; SD, 234.3 days), second-degree high ankle sprains (116.1 days; SD, 108.8 days), rotator cuff tears (113.5 days; SD, 133.6 days), third-degree MCL sprains (101.5 days; SD, 79.9 days), and sports hernias (96 days; SD, 78 days) (Table 2).

Trends Over Time

The 2022-2023 season had the greatest number of injuries (n = 133), while 2012-2013 had the fewest (n = 77). The overall number of injuries per year remained relatively consistent over all 14 seasons (Figure 1). However, diagnoses of concussions (see Figure 5A), intra-articular hip injuries (Appendix Figure A1, available in the online version of this article), and low back pain (Appendix Figure A1, available online) had relative increases over the study period. Across the different season sections there were 1206 (76.1%) regular season, 177 (11.2%) preseason, 119 (7.5%) postseason, and 83 (5.2%) off-season injuries.

Figure 1.

Number of injuries to National Hockey League goaltenders over 14 seasons. The slope represents the rate of change in the number of injuries.

Figure 5.

Summary of (A) number of concussion injuries per year (the slope represents the rate of change in the number of injuries per year), (B) time when the injury occurred, and (C) injury mechanism.

Injury Mechanisms

Most injuries were acute (n = 1511; 95.3%), and the most common injury mechanism was noncontact (n = 673; 42.5%), followed by incidental contact (n = 252; 15.9%), hit by puck (n = 118; 7.4%), and contact with the environment (n = 62; 3.9%). A total of 77 injuries (4.9%) resulted from intentional contact with another player, fighting, or body checks. Off-ice injuries accounted for the smallest proportion (n = 29; 1.8%). The remaining injuries (n = 374; 23.6%) did not have a listed mechanism or were classified as “other” (Figure 2).

Figure 2.

Percent of total number of injuries classified by injury category. This is generally representative of the injury mechanism.

Injury Timing

The majority of the injuries occurred during the regular season (all injuries n = 1206, 76%; game-only injuries n = 850, 88.3%). However, the EAIR was greater in pre- and postseason play (19.2 and 25.2 injuries/1000 athlete-exposures, respectively) (Table 3).

Table 3.

EAIR in the Preseason, Regular Season, and Postseason ^a

Time of Season	Total Injuries	Games Played	EAIR (/1000 athlete -exposures)
Preseason	49	1276	19.2
Regular season	850	33,912	12.1
Postseason	64	1272	25.2

^aOne exposure is equal to the number of games played by goaltenders multiplied by 2 (2 on ice at a time). EAIR, exposure-adjusted injury rate.

Among game-related injuries, the highest proportion occurred during the first period (n = 383; 24.2%), followed by an even distribution in the second (n = 232; 14.6%) and third (n = 228; 14.4%) periods (Figure 3). Pregame warm-up and overtime accounted for 2.8% (n = 44) and 1.0% (n = 17), respectively. Non–game-related injuries accounted for 43.0% (n = 681) of the total included cases. The highest proportion of non–game-related injuries were sustained in practice/scrimmage (n = 489; 30.9%), followed by off-ice (n = 98; 6.2%) and unknown (n = 94; 5.9%).

Figure 3.

Percent of total number of injuries classified by the period in which the injury occurred.

Contusions and other nonspecific injuries accounted for an additional 601 total injuries, resulting in 4117 TDM (Appendix Tables A1 and A2, available online). Contusions to the hand and knee were most common, as were nonspecific hip and knee injuries. These injuries generally resulted in no days missed, with a few potential outlier cases (eg, a knee contusion case resulting in 174 TDM) (Appendix Table A1, available online).

Proximal Adductor Injuries

Proximal adductor injuries were the most commonly occurring injury in the cohort (23.4% of injuries), and the number of proximal adductor injuries was relatively stable year over year (Figure 4A). Proximal adductor injuries represented a spectrum of pathology that included both strains (n = 328) and tears (n = 11) that resulted in a wide range of TDM from 16.2 (SD, 27.2) to 65.2 (SD, 45.9), respectively (Table 2). While a large proportion of these injuries (30.7%) occurred during practice, those that occurred in game tended to happen within the first period of play (28.0%) (Figure 4B). Most proximal adductor injuries occurred via a noncontact mechanism (n = 286; 77.1%) (Figure 4C).

Figure 4.

Summary of (A) number of proximal adductor injuries per year (the slope represents the rate of change in the number of injuries per year), (B) time when the injury occurred, and (C) injury mechanism.

Concussion

Concussions were the second most common injury (6.9% of all injuries) (Table 2), with a relative increase in their occurrence over the last 14 seasons (Figure 5A), resulting in a mean of 25.0 TDM (SD, 63.0 TDM). While there was an equal distribution of injuries between periods, a large percentage of concussions (n = 30; 27.3%) still occurred during practice (Figure 5B). The 2 primary mechanisms were incidental contact with another player (41.8%) and being struck by the puck (36.4%) (Figure 5C).

Medial Collateral Ligament Injury of the Knee

Medial collateral ligament (MCL) injuries were the third most frequently occurring among NHL goaltenders (6.0% of all injuries) (Table 2), with a relatively small increase over the 14-year period (Figure 6A). First-degree injuries were most common (n = 70), resulting in a mean of 18.7 TDM (SD, 27.2 TDM) (Table 2). Second-degree injuries occurred less frequently (n = 23) and resulted in 71.4 TDM (SD, 56 TDM) on average (Table 2). Third-degree injuries were uncommon (n = 2), with 1 player missing 45 days and the other missing 158 as a result of their injury. The majority of MCL injuries occurred during gameplay (85.3%) (Figure 6B), and there was an even distribution between noncontact (41.1%) and incidental contact (42.1%) with another player as the primary cause for injury (Figure 6C). A summary of the injury mechanisms and the period of injury for the remaining top 20 diagnoses are available in Tables 4 and 5, respectively.

Figure 6.

Summary of (A) number of medial collateral ligament injuries per year (the slope represents the rate of change in the number of injuries per year), (B) time when the injury occurred, and (C) injury mechanism.

Table 4.

Summary of Injuries That Occurred by Different Injury Mechanisms for the Remaining Top 20 Injuries (Rank 4-20) ^a

Diagnosis	Noncontact	Unknown/ Other	Incidental Contact With Other Player	Hit by Puck	Intentional Contact With Other Player	Contact With Environment	Off-Ice Injury
Cervical strain/whiplash	9	10	44	2	26	0	2
Low back pain	45	36	2	0	1	0	0
Intra-articular hip injury	32	31	6	0	2	1	1
Hamstring injury	24	15	5	0	0	3	0
Meniscal tears	20	15	2	0	0	1	1
Lacerations	2	2	12	19	2	2	0
Rotator cuff injury	11	12	5	0	1	6	3
Lateral ligament sprains	15	5	6	0	0	5	1
Abdominal wall injuries	18	8	3	0	1	0	2
Finger fractures	0	0	1	27	0	0	0
High ankle sprain	9	3	4	0	1	9	1
Paralumbar muscle strain	14	9	1	0	0	0	3
Patellar tendinopathy	16	9	0	0	0	0	0
Iliopsoas strain	13	10	1	0	0	0	0
Wrist sprain	5	4	6	2	1	4	0
Elbow UCL sprain (acute)	1	1	12	2	1	3	0
Quadriceps strain	16	1	1	0	0	0	0

^aUCL, ulnar collateral ligament.

Table 5.

Summary of Injuries That Occurred by Different Periods of Play for the Remaining Top 20 Injuries (Rank 4-20) ^a

Diagnosis	Pregame/ Warm-up	First Period	Second Period	Third Period	Overtime	Practice/ Scrimmage	Off-Ice	Unknown
Cervical strain/whiplash	1	27	28	17	3	10	5	2
Low back pain	3	20	8	7	0	28	11	7
Intra-articular hip injury	1	17	7	7	1	29	5	6
Hamstring injury	3	8	9	9	0	13	3	2
Meniscal tears	1	4	6	3	1	18	2	4
Lacerations	1	12	5	5	0	14	0	2
Rotator cuff injury	0	15	5	3	1	8	4	2
Lateral ligament sprains	2	5	4	7	0	8	4	2
Abdominal wall injuries	2	12	5	1	1	5	3	3
Finger fractures	3	3	1	1	0	20	0	0
High ankle sprain	0	7	4	8	0	6	1	1
Paralumbar muscle strain	1	2	5	5	0	7	5	2
Patellar tendinopathy	0	2	1	2	0	11	2	7
Iliopsoas strain	2	6	3	0	1	9	1	2
Wrist sprain	2	7	1	5	0	6	0	1
Elbow UCL sprain	1	5	2	7	0	3	1	0
Quadriceps strain	1	6	3	4	0	4	0	0

^aUCL, ulnar collateral ligament.

Discussion

The purpose of this study was to quantify the number of injuries occurring in NHL goaltenders over the last 14 seasons and to determine the time lost as a result of each category of injury. The primary findings of this investigation demonstrate that proximal adductor injuries are the most frequently occurring, followed by concussions and MCL sprains. The overall injury incidence was consistent year over year and was greater in the pre- and postseasons compared with the regular season. The highest proportion of injuries occurred by a noncontact mechanism, and the majority of the injuries occurred during the regular season.

Injury risk reduction is a significant priority in professional sports, with teams investing substantial resources into protecting their players. Injury risk reduction remains particularly important for ice hockey goaltenders, as these athletes represent significant economic investments for their teams, with fewer replacement options available in the event of injury. In the 2023-2024 season, NHL goaltending tandems comprised 8.17% (3%-15%) of a team’s salary cap, on average. ³⁰ Injuries to goaltenders at the elite level tend to result in longer periods of time away from play, compared with positional players, at both the NHL ²⁶ and National Collegiate Athletic Association (NCAA) levels. ³⁹ At the NHL level, being a goaltender is a risk factor for an injury that results in >5 man-games lost. ²⁶ To our knowledge, this is the largest data set using injury data directly from the NHL AHMS.

Hip and Proximal Adductor (Groin) Injuries

Hip and groin injuries are common in hockey players.^23,27 In a systematic review, Berube et al ⁵ identified that hip and proximal adductor injuries accounted for 10% and 8% of injuries in NHL hockey players, across all positions, respectively. Goaltenders were identified as having a greater risk of intra-articular injury (eg, femoroacetabular impingement syndrome) compared with forwards and defensemen. ⁵ In a prospective study of European professional goaltenders, 69% reported at least 1 problem of the hip or groin, with 36% experiencing an injury that they considered substantial. ⁴¹ In our cohort of only goaltenders, there were fewer hip injuries (4.6%) but more adductor injuries (23.4%) than those previously reported in general hockey populations. Of note, these injuries were one of the few in our cohort that appeared to be increasing year over year. Although this is likely multifactorial, better recognition and reporting of these injuries is likely a significant contributor to the increased diagnoses.^3,23 The differences noted between goaltenders and position players highlight the unique demands and injury patterns experienced by goaltenders and the need for goaltender-specific research and injury risk reduction interventions.

Modern goaltending styles utilize the butterfly technique (Figure 7a), characterized by the goaltender dropping to their knees with their pads flared out to cover the lower part of the net. ¹⁴ This has become the predominant style of play due to its effectiveness in blocking low shots and for the ease of transitioning from this position into lateral movements. ⁴⁰ The popularity of the butterfly technique, however, may align with the high rates of noncontact injuries to the hip and groin in this cohort. Groin injuries occur due to eccentric activation of the proximal adductor muscles, often during high velocity and high force leg abduction, creating an excessive passive stretch.^6,33 While in the butterfly position, transitional movements such as lateral slides, push-off from the post, and kick or sprawling saves put the proximal adductor muscles at risk of injury from this mechanism. ¹⁷ Observational studies have shown that these commonly performed movements for NHL goaltenders are also perceived as being the most physically demanding ⁴² and are performed, on average, 34 times each game. ⁴ These repetitive motions about the hip are likely precursors to hip injuries in hockey goaltenders and partially explain the relatively high incidence of injuries observed in these athletes. ³⁸

Figure 7.

Examples of a goaltender in the (a) butterfly, (b) vertical-horizontal, and (c) reverse vertical-horizontal positions. The image was adapted from Wörner et al. (2021), ⁴² under the Creative Commons Attribution-Non-Commercial-No Derivatives 4.0 International license. The image has been modified to exclude a fourth position.

The results presented in the current investigation, specific to ice hockey goaltenders, are also consistent with other professional sports. ⁷ Hip and groin injuries are also one of the most common injuries in the National Basketball Association, the National Football League, and professional European football leagues. ⁷ In a more recent study, lower extremity injuries were the third most common type of injury, and strains and sprains resulted in the most days missed (only behind fractures) in hockey athletes. ¹⁹ However, this study was generalized to all hockey players and did not distinguish between goaltenders and position players. ¹⁹ Despite the high rate of injuries to goaltenders presented in the current study, there is minimal literature on the epidemiology of injuries to goaltenders at the professional level.

Concussions

Awareness surrounding head injuries has grown steadily in the NHL over the last 2 decades.^16,34 Despite this increase in awareness, concussions represented the second most common injury among our cohort and, consistent with other recent ¹⁶ reports, have remained steady over the last 20 years. Andrews et al ² analyzed the return-to-play times after concussions in NHL position players and demonstrated a mean return to play of 17.2 days. Our results suggest that return-to-play times in goaltenders (25 TDM or 16.8 TDM when outliers removed) are consistent with those of position players. While there has also been advancement in concussion treatment and rehabilitation, ³⁷ they are still associated with decreased performance metrics up to 3 years after injury when compared with the year before the concussion. ²

Ice hockey goaltenders face a distinct vulnerability to concussions in the form of impacts from the puck. In the current study, 36.4% of concussions were sustained via this mechanism. Additionally, goaltenders experience collisions with other players, and any type of player contact (incidental and intentional) accounts for >60% of the concussions in the current data set. Mechanical testing on common helmet configurations utilized in NHL goaltenders revealed that a helmet’s ability to reduce the parameters associated with concussion is dependent on the type of helmet (ie, helmet stiffness and liner thickness).^8,9 In addition, the ability of a helmet to reduce the risk of puck impact–related concussion is highly dependent on the location of the impact.^8,9 Finally, studies that evaluate the effectiveness of helmets in reducing the risk of concussion have not thoroughly considered the subconcussive impacts sustained by goaltenders and the cumulative brain trauma that this can produce.^24,28 Therefore, there remains a need for research into goaltender helmet design to further reduce the risk of concussion. In addition, improvements to head protection may also be warranted during practice, as this was the setting of 27.3% of the concussions in the current investigation.

MCL Injury of the Knee

MCL injuries were the third most common injury and accounted for the greatest TDM, underscoring their severity and impact on player availability. The majority of injuries were low grade, resulting in a mean of 18.7 days missed. Intermediate- and high-grade sprains occurred far less frequently but were associated with significantly more morbidity, resulting in 71.4 and 101.5 days missed, respectively. The butterfly style of play may also influence the occurrence of MCL injuries experienced by goaltenders. Injuries to the MCL typically occur when the knee is in a position of extreme valgus, and the risk is further increased when an external valgus force is applied.^22,32,44 During the butterfly, the maximum knee valgus angle of the lower extremities has been quantified as approximately 11° (an at-risk position ¹⁸ ), and the knee tends to maintain this position throughout the task. ²⁰ A likely in-game scenario may occur when a position player comes into contact with the goaltender while they are making a play or falling in the crease, as incidental contact accounted for 42.1% of MCL injuries.

Equipment choice may also influence the risk of MCL injury. Modern leg pads utilize a leg channel where the athlete’s knee, leg, and ankle are affixed with a variety of straps, laces, and buckles, forming a mechanical link between the pads and the lower extremity. ¹⁵ Frayne and Dickey ¹⁵ quantified the motion of the pads relative to the lower extremity and found that a stiff but wide leg channel produced significantly greater transverse plane rotation (ie, external rotation of the pad relative to the leg) compared with 3 other pad conditions. They suggest that the overrotation of the pad could place the lower extremity in a compromised position, potentially increasing the risk of injuries to the knee. ¹⁵ In addition to the type of pads that are used, the individual setup of the pads, which is specific to goaltender preference, ⁴⁰ may produce different force patterns acting on the lower extremity. As the lower extremity behaves as a coupled kinematic chain,^10,11 altering or prohibiting motions at one joint will have an effect on the biomechanics at adjacent joints. ³⁵ For example, if leg pads are tightly fastened, this may force the ankle into a pronated posture when the goaltender drops into the butterfly position or pushes off from the post. Pronation of the ankle and subtalar joint has been shown to increase knee abduction moment, hip internal range of motion, ³⁶ and knee internal rotation angle. ³¹ The increases in knee abduction moment and hip internal rotation are consistent with dynamic knee valgus and therefore indicate a mechanism for increasing the demand on the MCL.^21,29 Future research, therefore, should focus on optimizing pad fit and leg channel setup to account for individualized anatomy as an intervention to reduce the risk of injury while maintaining a high level of performance.

Injury Trends and Timing

Overall, the yearly incidence of injuries has remained consistent over the last 14 seasons, with the exception of concussions and intra-articular hip injuries, which appear to be increasing year to year. This is despite the introduction of 3 separate rule changes introduced over the course of the injury reporting that decreased goaltender protective padding. These changes included a mandated decrease in leg pad width from 12 inches to 11 inches in 2009-2010 and a decrease in pant and chest protector bulk in the 2017-2018 and 2018-2019 seasons, respectively.

In the current study, the greatest number of total injuries occurred during the regular season, while the EAIRs were higher in the pre- and postseason. This suggests that goaltenders may be at a higher risk of injury during these periods. Previous research has shown relatively inconsistent findings related to time of season and incidence of injuries. Doyle et al, ¹³ in their systematic review and meta-analysis, indicated that there were no differences in anterior cruciate ligament injury rates between the first and second half of a season, but there was a trend toward an increased risk of groin injuries in the first half of a season compared with the second. ¹³ Other research has identified similar results to the current investigation, where the injury rates in women’s collegiate soccer players were higher in the preseason and postseason compared with the midseason injury rates. ¹² While there is little consensus on why the time of season may affect injury rates, we hypothesize that in the preseason, players may not yet be fully adjusted to the speed and physicality of the game. In the postseason, the increased rate of injury may be due to the condensed schedule and increased game intensity.

Lastly, a notable finding in our study was the number of injuries that occurred during practice. In our cohort, this represented 30.9% of all injuries, with certain injuries, such as proximal adductor injury, intra-articular hip injury, concussion, and low back pain, having a significant percentage of events occurring during non-gameplay. This is consistent with previous work that also found a relatively high incidence rate of injuries in men’s and women’s NCAA collegiate hockey players at 4.7 and 4.2 injuries per 1000 athlete-exposures during preseason practices, respectively. ¹ Although NHL goaltenders face 25 to 40 shots and perform the butterfly movements approximately 34 times during a typical game, ⁴ this number is likely higher during practice. The number of concussions occurring during practice is alarming but unsurprising given the volume of shots being faced. Further strategies to protect goaltenders during practice, such as limiting shot numbers or using extra-padded equipment, may be warranted but would require further study.

Limitations

While this is the largest descriptive study on injuries to professional goaltenders, it is not without limitations. This was a retrospective study using a third-party database, and as such, we were limited in the scope of data that were provided, which was at the discretion of the NHL. The restricted demographic data, implemented to safeguard anonymity, hinder a comprehensive understanding of individual player or equipment factors influencing injuries. Furthermore, it was not possible to analyze in-game scenarios that may play a role in injury (eg, number of shots faced, time on penalty kill, and games played on a back-to-back night). Future studies should strive to incorporate such data where possible, to elucidate the relationship between player or game characteristics and injury occurrence to enhance the efficacy of injury risk reduction strategies. Lastly, with the exclusion of treatment information, we could not make a determination on which injuries required or benefitted from medical or surgical treatment and how that influenced return to hockey activities.

Conclusion

The most common injuries to NHL goaltenders over the last 14 seasons were proximal adductor injuries, concussions, and MCL sprains. The overall injury incidence has been consistent and is greater in the pre- and postseasons compared with the regular season, with a large percentage of injuries occurring during practice. Understanding injury patterns and the relative timeline of return to play for goaltender-specific injuries on a granular level may allow team physicians to better manage player expectations and could have a significant effect on team decisions regarding roster management. Future studies aimed at targeted injury reduction strategies are warranted and should incorporate demographic and game-exposure data where possible.