Table 3.

Studies reporting on the “Size Classification”

Study	Patients/ ankles	Sex/ age	Imaging	Description	PMF type distribution	Treatment allocation	Cut-off	Assesing Classification	Reliability		Predictive outcome values
									Inter OR	Intra OR
Warner (1965) [37]	n.r./100	n.r	x-ray	Percentage of tibial articular surface	Unclear	0–20%: closed reduction, 20–30% only fix MM/LM, 30–60%: ORIF; via transfibular approach	≥ 20%	No	n.r	n.r	n.r
McDaniel (1977) [58]	50/51	13m, 37f, 41 (20–68) yo	x-ray	Percentage of tibial articular surface	< 10% (9), 10–25% (27), ≥ 25% (15)	Closed methods (28), treated operatively (23; 7 screw fixation PMF ≥ 25%)	≥ 25%	Yes	n.r	n.r	In PMF ≥ 25% ORIF associated with better outcome than closed treatment
Yde (1980) [99]	487/488	263m, 224f, 44 yo	x-ray	Percentage of joint surface	SER3: Only 14% of cases PMF ≥ 25%; PER4: In 73% of cases PMF ≥ 25%	n.r	≥ 25%	No	n.r	n.r	n.r
Olerud (1986) [38]	134/134	Not clear/48 (15–86) yo	x-ray	Percentage of joint surface	86 of 134 with PMF, < 1⁄6 of surface (49), 1⁄6 -1/3(32), ≥  1/3(5)	Posterior fragments either left alone or fixed with screw	No cut- off	No	n.r	n.r	PMF without displacements best
Jaskulka (1989) [4]	142/142	47m, 95f, 46.7 yo	x-ray	Percentage of articular surface	62 of 142 with PMF: avulsion of posterior Lip (32), PMF 5–20% (15), PMF ≥ 20% (15)	14 ORIF with screws (none of the posterior lip); if < 20% generally no osteosynthesis	No cut- off	Yes	n.r	n.r	Longterm outcome poorer if fracture of posterior margin present (even small fragments); poorest results: conservativly treated PMF
Broos (1991) [8]	604/612	320m, 284f, 41 yo	x-ray	One-third of the joint surface	178 of 612 fractures with PMF	100 of 178 patients with PMF fixed with screws; PMF < 1/3 left untouched	≥ 1/3	Yes	n.r	n.r	If PMF present only 66% excellent or good result compared with 81% without PMF; ankle fx with PMF > 1/3 with perfect anatomic fixation, give worse results than ankle fractures with small unfixed PMF
Broos (1992) [100]	604/612	320m, 284f, 41 yo	x-ray	One-third of the joint surface	178 of 612 fractures with PMF	100 of 178 with PMF fixed with screws; PMF < 1/3 left untouched	≥ 1/3	Yes	n.r	n.r	In spite of perfect anatomic internal fixation of fragments > 1/3, the results remained worse than with small non fixed posterior fragments
Heim (2002) [39]	n.r./111	n.r	x-ray	One-third of the articular surface	n.r	Surgically fixed if PMF ≥ 1/3 size	≥ 1/3	No	n.r	n.r	n.r
Langenhuijsen (2002) [7]	57/57	21m, 36f, 49.6 (18–89) yo	x-ray	Percentage of articular surface	PMF < 10% (24); 10–25% (19), > 25% (14)	53 of 57 treated operatively, 4 conservatively; ORIF (14) (1x < 10%, 3 × 10–25%, 10x ≥ 25%)	No cut- off	Yes	n.r	n.r	Size or fixation of fragment did not influence prognosis; joint congruity was significant factor for prognosis, congruity should be achieved for fragments > 10% of the tibial articular surface
Papachristou (2003) [101]	15/15	4m, 11f, 48.8 (14–73) yo	x-ray	Perecntage of articular surface	All between 1/4—1/3 of articular surface	All ORIF PL or PM approach	≥ 25%	No	n.r	n.r	n.r
De Vries (2005) [52]	45/45	10m, 35f, 61 (37–81) yo	x-ray	Percentage of articular surface	PMF < 25% (32), PMF > 25% (13)	11 of 45 patients with PMF fixation: PMF ≥ 25% (8), PMF < 25% (3)	≥ 25%	Yes	n.r	n.r	PMF fixation did not have better outcome than not fixated (AFSS); no evidence for the need for fixation of fragments smaller than 25%
Haraguchi (2006) [14]	57/57	41m, 16f, 43 (13–80) yo	CT	Percentage of tibial plafond	9 of 57 Fx ≥ 25% (7 of 9 extended to medial malleolus)	All fractures treated with ORIF, best approached medially because of medial extension	≥ 25%	No	n.r	n.r	n.r
Farsetti (2009) [102]	44/44	19m, 25f, 43 (18–69) yo	x-ray	Percentage of joint surface length	26 trimalleolar fractures	ORIF if PMF > 25% (10)	≥ 25%	No	n.r	n.r	n.r
Büchler (2009) [13]	22/22	7m, 15f, 50 (21.9–72.2) yo	x-ray + CT	Percentage of articular surface	All trimalleolar fractures, 19 ankles (86%) with a PM lesion of tibial plafond	n.r	No cut- off	Yes	Kap-pa 0.64	Kappa 0.63	n.r
Forberger (2009) [103]	45/45	12m, 33f, 54 (18–83) yo	x-ray	Percentage of articular surface	PMF ≥ 25% (19); PMF ≥ 10% surgically treated if younger (under 50yo) or with subluxation (n = 26)	PL approach used for osteosynthesis in 45 patients	≥ 25%  ≥ 10% if sub-luxation or < 50 yo	No	n.r	n.r	n.r
Tejwani (2010) [6]	309/n.r	n.r./43.3 yo	x-ray	Percentage of fragment size	54 of with PMF	Fixation decision by surgeon based on size and tibio-talar stability: fixation (20) average size 25.2%; not fixed (34) average size 16.1%	No cut- off	No	n.r	n.r	Trend for better outcome scores for patients who had fixation (numbers too small to meaningfully compare fixation versus no fixation groups)
Hong-Chuan (2010) [104]	96/96	75m, 21f, 38 (17–79) yo	x-ray	Percentage of articular surface	PMF < 25% (10); PMF ≥ 25% (26)	PMF ≥ 25 fixed (4 × PM approach, 3 × PL approach, 19 × medial approach; PMF < 25% reduced spontane. after reduction of LM	≥ 25%	No	n.r	n.r	n.r
Heim (2010) [77]	43/43	15m, 28f, 58.8 yo	x-ray	Percentage of articular surface	No PMF (7), small shelllike (10), PMF < 25% (18), PMF ≥ 25% (8)	8 PMF ≥ 25% fixed: AP screw (4), PA screw (1), plate (3)	No cut- off	Yes	n.r	n.r	Volkmann fx worse prognosis; fixation with anatomic results does not return score to norm (OMAS), instability as indicator for PMF fixation
Mingo-Robinet (2011) [46]	45/45	10m, 35f, 50.69 ± 17.32 (20–80) yo	x-ray	Percentage of articular surface	PMF < 25% (20), PMF ≥ 25% (25)	Ligamentotaxis (27), AP fixation (15), PL fixation (3)	No cut- off	Yes	n.r	n.r	Tendency toward worse result when PMF ≥ 25%; better radiological results when small PMF reduced; results encourage to fix all PMF ≥ 25% and fix most of smaller PMF not satisfactorily reduced by ligamentotaxis
Purnell (2011) [105]	66/67	45m, 21f, 44 (18–69) yo	CT	Percentage of articular surface	23 of 67 with PMF	PMF fixed if ≥ 25% (AP screws after closed reduction)	≥ 25%	No	n.r	n.r	n.r
Abdelgawad (2011) [70]	12/12	10m, 2f, 41 (20–61) yo	x-ray	Percentage of tibial plafond	All ≥ 30% of tibial plafond	Fixation of PMF done if ≥ 30% of tibial plafond and more than 2 mm displaced after closed reduction (12) with PL approach	≥ 30%	No	n.r	n.r	n.r
Xu (2012) [53]	102/102	41m, 61f, 43.4 (15–80) yo	x-ray	Percentage of articular surface	PMF < 10% (16), PMF 10–25% (62), PMF ≥ 25% (24)	42 of 102 surgically fixed; PMF < 25% fixed (18) unfixed (60)	No cut- off	Yes	n.r	n.r	Arthritis score for different PMF sizes differed greatly; difference in treatment effect between articular surface evenness and unevenness for all fragment; no difference in treatment effect between fixed and unfixed PMF (AOFAS; VAS)
Di Giorgio (2013) [106]	18/18	12m, 6f, 44.2 (26–61) yo	x-ray	Percentage of articular surface	8 of 18 with PMF; ≥ 25% (8)	Fixating PMF ≥ 25% and ignoring smaller fragments followed in 82% of patients (21); in 5 of 13 (38%) PMF ≥ 25% not fixated	≥ 25%	No	n.r	n.r	n.r
Hoelsbrekken (2013) [107]	82/82	31m, 51f, 52.8 (± 15,3) yo	x-ray	Percentage of articular surface	n.r	ORIF of PMF if the fragment ≥ 25% of the articular surface	≥ 25%	No	n.r	n.r	n.r
Hong (2014) [109]	31/31	10m, 21f, 46 (20–76) yo	x-ray	Percentage of articular surface	PMF < 25% (21), PMF ≥ 25% (10); mean size 22.3% (range 6% to 46%)	8 of 31 PMF fixed (≥ 25%)	≥ 25%	No	n.r	n.r	Poorer functional outcome with increasing posterior malleolar fragment size (OMAS score)
Erdem (2014) [108]	40/40	20m, 20f, 48.9 (24–51) yo	x-ray + CT	Percentage of articular surface	n.r	Surgical treatment if fracture > 20%, screws (20), plate (20)	≥ 20%	No	n.r	n.r	n.r
Evers (2015) [47]	42/42	16m, 26f, 52.8 (19–86) yo	x-ray + CT	Percentage of joint surface involvement	Group I: 54,8% < 25% (23), group II: 45% ≥ 25% (19)	< 25% osteosynthesis (2), ≥ 25% osteosynthesis (14)	No cut- off	Yes	n.r	n.r	Patients with PMF fixation had higher probability achieving good AOFAS Score and therefore improved outcomes
Kim (2015) [110]	36/36	16m, 20f, 58 (23–85) yo	x-ray	Percentage of articular surface	All ≥ 30% of the tibial plafond	All lateral transmalleolar approach and miniscrews fixation	≥ 30%	No	n.r	n.r	n.r
Verhage (2015) [111]	243/243	114m, 129f, 52 yo	x-ray	Percentage of articular surface	Mean PMF size 16% (range 3–53%)	Fixation in 11 cases: 8x ≥ 25%, 3 × 5–25%, large posterior fragments reduced closed or percutaneously and fixed by 1- 2 AP screws	No cut- off	Yes	n.r	n.r	Fragments > 5% resulted in worse functional outcome than fragments < 5% (not statistically significant)
Choi (2015) [112]	50/50	27m, 23f, 47.5 (25–75) yo	CT	Percentage of articular surface	Mean area 31.7%	ORIF with single oblique PL approach; included involvement of PMF < 25%	No cut- off	No	n.r	n.r	n.r
Drijfhout van Hooff (2015) [71]	131/131	55m, 76f, 51 (24–74) yo	x-ray	Percentage of joint surface	Small (< 5%, n = 20), medium (5–25%, n = 86), or large (≥ 25%, n = 25)	In 24 patients PMF fixed (mainly percutaneously in AP direction)	No cut- off	Yes	n.r	n.r	No difference in outcome comparing small PMF to medium and large fragments (AOFAS, AAOS, VAS,RoM); patients with PMF > 5% developed more OA; anatomical fixation of PMF between 5 and 25% may lead to improved outcome
Endo (2016) [113]	20/20	15m, 5f, 47 (18–67) yo	CT	Percentage of articular surface	n.r	PMF ≥ 25% fixed with 1—2 screws; PMF < 25% of the articular surfaces were fixed depending on surgeons preference	No cut- off	No	n.r	n.r	n.r
Chan (2016) [114]	254/254	106m, 148f, 45.1 (13–83) yo	x-ray + CT	One-third of the articular surface	95 of 256 with PMF	Only 13 with fixation (fixed with guide pins through PL approach if displaced > 2 mm and involved > 1/3 size)	≥ 1/3	No	n.r	n.r	n.r
Guo (2017) [54]	284/284	156m, 128f, 39.12 (± 12.23) yo	CT	Percentage of joint surface involve-ment	< 25% (239), ≥ 25% (44)	Fixation ≥ 25% (40) < 25% (72) AP screws; no fixation ≥ 25% (4) < 25% (168) conservatively	No cut- off	Yes	n.r	n.r	PMF < 25%: no difference in outcomes fixed vs. not fixed in tibial spiral fx; PMF > 25% significantly improved clinical outcome with fixation (AOFAS; VAS)
Naumann (2017) [115]	1011/1011	455m, 556f, 51.4 (18–94) yo	x-ray	≥ ¼ of the articular surface	235 of 1011 with PMF	ORIF if PMF ≥ 1/4 of the articular surface	≥ 25%	No	n.r	n.r	n.r
Vidovic (2017) [116]	46/46	28m, 18f, 52.3 (33–71) yo	CT	Percentage of articular surface	All ≥ 25% of articular surface	Indirect reduction and percutaneous AP fixation (22); direct reduction and PA fixation (24)	≥ 25%	No	n.r	n.r	Significantly better quality of reduction with direct reduction
Shi (2017) [117]	116	59m, 57f, 48.6 (± 13.36) yo	x-ray + CT	Percentage of articular surface	All ≥ 25% of articular surface	Direct fixation (plate or screw) (64), indirect fixation (ligamentotaxis) (52)	≥ 25%	No	n.r	n.r	Direct reduction technique provide better quality of fracture reduction and outcome of PMF > 25% (AOFAS)
Saygili (2017) [55]	73/73	34m, 39f, 42.65 (± 13.68) yo	x-ray	Percentage of joint surface	All < 25% of joint surface	PMF fixation (10 × screw, 17 × plate), no fixation (46)	≥ 25%	Yes	n.r	n.r	PMF fixation vs no fixation had no significant difference (AOFAS)
Zhou (2017) [119]	34/34	15m, 19f, 41.2 (21–71) yo	3DCT	Percentage of distal articular tibia	All ≥ 25% of articular surface (1/4–1/3(19), 1/3–1/2 (11), ≥ 1/2 (4))	ORIF: plates and screws via a PL approach	≥ 25%	No	n.r	n.r	n.r
Wang (2017) [120]	84/84	39m, 45f, 44.16 (15–75) yo	CT	Percentage of articular surface	40 of 84 patients with PMF ≥ 25%	PMF ≥ 25% fixed with 2 screws via PL approach; PMF remain untreated when fracture < 25%	≥ 25%	No	n.r	n.r	n.r
Xing (2018) [121]	30/30	19m, 11f, 30 (22–65) yo	x-ray + CT	Percentage of articular surface	All ≥ 25% of articular surface	PMF fixed with screw or Kirschner wire	≥ 25%	No	n.r	n.r	n.r
Baek (2018) [122]	29/29	20m, 9f, 31.3 (17–55) yo	x-ray + CT	Percentage of articular involve-ment	19 of 29 with PMF	n.r	≥ 20%	No	n.r	n.r	71.4% of the patients with malreduced syndesmosis had associated PMF without fixation since < 20% size
Levack (2018) [123]	178/178	63m, 115f, 50.2 (16.9) yo	CT	Percentage of 2-D tibial plafond area	122 of 178 with PMF: 10% (60%), 10%–20% (28%), ≥ 20% (12%)	PMF reduced and with plate regardless of fracture size	No cut- off	No	n.r	n.r	n.r
Kim (2018) [124]	19/19	7m, 12f, 45.1 (13–68) yo	x-ray	Percentage of articular surface	All ≥ 25% of articular surface	Release of PITFL through PL approach for PMF reduction with posterior screws (19)	≥ 25%	No	n.r	n.r	n.r
Tosun (2018) [51]	49/49	21m, 28f, 47 (20–82) yo	CT	Percentage of articular surface	Mean size of PMF: 21.3% in group 1, 28.9% in group 2	29 × untreated PMF; group 2: PMF treated regardless size (surgeons preference) 8 of 20 PMF < 25% fixation: screw or plate	No cut- off	Yes	n.r	n.r	Recommend all PMFs to be fixed regardless of size (related to successful radiological and functional outcomes, AOSAF, OA)
Miller (2018) [125]	198/198	69m, 129f, 48.86 (± 15,98) yo	x-ray	Percentage of articular width	Mean in prone group (n = 47): 22.2%, mean in supine group (n = 151): 15.2%	Prone: 100% fixation (46 × plate, 1 × syndesmotic); supine: 24.5% fixation (37) (2 × plate, 41 × syndesmotic)	No cut- off	No	n.r	n.r	n.r
Huang (2018) [126]	111/111	72m,39f, 46.6 ± 16.1 yo	CT	Percentage of articular surface	42 of 111 with PMF	Surgical treatment (19) PMF ≥ 25%, conservative treatment (23)	≥ 25%	No	n.r	n.r	n.r
Xing (2018) [127]	69/69	41m,28f, 37.36 (20–67) yo	x-ray + CT	Percentage of articular surface	All at least 25% of the articular surface	PMF anatomically reduced by screw	≥ 25%	no	n.r	n.r	n.r
Baumbach (2019) [48]	236/236	99m, 137f, 53.0 (± 18.3) (18–100) yo	CT	Percent tibial plafond depth	< 25% (169), ≥ 25% (67)	ORIF (48 < 25%, 30 > 25%), CRIF (14 < 25%, 30 > 25%), none (107 < 25%, 7 > 25%)	No cut- off	Yes	n.r	n.r	All PMF independent of size should be treated by ORIF, as this restores syndesmotic stability significantly more often than untreated PMF
Kang (2019) [49]	62/62	35m, 27f, 50.7 (19–78) yo	CT	Percentage of tibial joint surface	All affecting > 10%—< 25% of the tibial joint surface	32 × screw fixation; 30 × internal fixation malleolar fractures and no screw fixation PMF	No cut- off	Yes	n.r	n.r	Surgical fixation of PMF < 25% results in better clinical outcome (AAOS, FS-36, FAOS)
Meijer (2019) [132]	104/104	41m, 63f, 63 (37–91) yo	x-ray	Percentage of articular surface	n.r	25% cut-off; fixation of PMF in 19 fixated (18%) with AP or PA screw	≥ 25%	No	n.r	n.r	n.r
Meijer (2019) [131]	31/31	12m, 19f, 46 (19–73) yo	Q3DCT	Percentage of articular surface	Mean articular surface of 12% (range 0.0%–36.0%) of the joint surface	All ORIF: direct fixation (14), syndesmotic screw (4), no fixation (13)	No cut- off	Yes	n.r	n.r	Fragment size was correlated with arthrosis; no significant correlations were found for FAOS and SF-36 scores
Testa (2019) [133]	48/48	22m, 26f, 44.7 (19–72) yo	x-ray	Percentage of articular surface	Mean fragment size 18%, mean fragment size fixing 26%, 13 of 48 ≥ 25%	Fixation of PMF in 8 of 48 when PMF ≥ 25%, fixed with screws	≥ 25%	No	n.r	n.r	No correlation between size of PMF and ankle dislocation and size of PMF and syndesmosis stability
Verhage (2019) [72]	169/169	67m, 102f, 52.3 yo	x-ray	Percentage of intra-articular surface	< 5% (20), 5–25% (119), ≥ 25% (30)	39 × Fixation of PMF: none of < 5% were fixated, 15 of the 5–25% fixated,24 of ≥ 25% fixated	No cut- off	Yes	n.r	n.r	Size and fixation of PMF not associated with OA or outcome; postoperative step-off as risk factor
Pilskog (2020) [56]	130/130	36 m, 94f, 57 (41–67) yo	x-ray	Percentage of joint involve-ment	Median PMF size 17%; 65 PMF < 25% (31 × fixation, 35 × no fixation)	PMF fixation in 42 of 43 (98%) in group A, 7 of 43 (16%) in group B	No cut- off	Yes	n.r	n.r	PMF < 25% fixed vs. not-fixed same SEFAS, RANS-36, VAS
Blom (2020) [63]	70/70	30m, 40f, 47 (SD 14.6) yo	CT	Percentage of tibial articular surface	Mean of 15.9% of articular surface	According to AO principles of ORIF	No cut- off	Yes	n.r	n.r	PMF morphology and not size found as predictor of FAOS score
Wang (2020) [138]	243/243	120m, 123f, 44.27 (± 12,5) yo	CT	FAR from CT crossectional images	≥ 15% (136) and < 15% (107)	All ORIF: reduction of PMF via PL approach, PA/ AP screws and PA plate used	No cut- off	No	n.r	n.r	n.r
McHale (2020) [50]	75/75	24m, 51f, 56.2 yo	x-ray	Percentage of fragment size	< 10% (34), 10–20% (18), 20–30% (12), ≥ 30% (11)	25 of 75 fractures fixed (surgeons preference) using PL approach (22 × plate, 3 × PA screw)	No cut- off	Yes	n.r	n.r	Worse outcome (MOXFQ, EQ-5D VAS) when PM size 10–20%, due to higher rate of anatomic reduction when fixed; best PROMs in PMF < 10%
Lee (2020) [134]	166/166	97m, 69f, 44.6 (18–83) yo	CT	Percentage of intra-articular surface	92 of 166 Fx with PMF (51 ≥ 25%)	Screw fixation PMF ≥ 25% and no long proximal extension; additional plate with PMF ≥ 25% and long proximal extension	≥ 25%	No	n.r	n.r	Incidence of chronic syndesmotic instability higher, regardless of PMF fixation
Liu (2020) [141]	85/85	48m, 37f, 42.92 (21–66) yo	not clear	Percentage of articular surface	All ≥ 25%; exclusion: medial malleolus extension type of PMF	plate group (n = 47), screw group (n = 38)	≥ 25%	No	n.r	n.r	n.r
Palmanovich (2020) [62]	85/85	23m,62f, 53.4 (17–89) yo	CT	1/3 of antero-posterior curvature of articular surface	All over 1/3 articular surface	Single PA screw fixation	≥ 1/3	No	n.r	n.r	n.r
Lee (2020) [139]	70/70	41m, 29f, 51.0 (18–78) yo	CT	Percentage of articular surface	n.r	PMF not fixed if no rotation or displacement, if PMF ≥ 20% fixation	≥ 20%	No	n.r	n.r	n.r
Ceccarini (2021) [144]	135/135	82m, 53f, 38.2 (19–50) yo (9 lost to FU)	x-ray	Percentage of articular surface	n.r	≥ 25% fixation with screw in AP or PA direction	≥ 25%	No	n.r	n.r	n.r
Schoen- makers (2022) [78]	26/26	10m, 16f, 58.08 ± 18.9 (22–78) yo	CT	Percentage of surface area	< 10% (6), 10%–25% (14), > 25% (6)	Surgical treatment 14 of 26 (9 × 10–25%, 4x > 25%)	No cut- off	Yes	n.r	n.r	Optimal value for PMF size to predict excellent AOFAS score was < 15%; PMF size ≤ 10% had a significantly greater RoM
Yamamoto (2022) [57]	110/110	44m, 66f, 52.5 ± 12.9 yo	CT	Percentage of intra-articular surface	< 25% ( 62), > 25% (48)	n.r	No cut- off	Yes	n.r	n.r	PMF < 25% fragment reduction, fixation, and ankle stability not associated with AOFAS score;
Pflüger (2022) [24]	193/193	n.r	CT	Proportion of PMF in relation to tibial diameter	0%-10% (11), 11%-15% (20), 16%-20% (30), 21%-25% (54), 26%-30% (36), 31%-35% (15), > 35% (27)	Surgical fixation PMF with larger size compared to without, no influence of size on direct or indirect fixation method	No cut- off	Yes	n.r	n.r	n.r