Monitoring fatigue state with heart rate‐based and subjective methods during intensified training in recreational runners

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ Moni o ing a igue s a e wi h hea a e‐based and subjec i e me hods du ing in ensi ied aining in ec ea ional unne s © 2024 The Au ho s. Eu opean Jou nal o Spo Science published by Wiley‐VCH GmbH on behal o Eu opean College o Spo Science Published e sion Nuu ila, Olli‐Pekka; Uusi alo, A ja; Kokkonen, Veli‐Pekka; Wee a a hna, Nilushika; Ky öläinen, Heikki Nuu ila, O., Uusi alo, A., Kokkonen, V., Wee a a hna, N., & Ky öläinen, H. (2024). Moni o ing a igue s a e wi h hea a e‐based and subjec i e me hods du ing in ensi ied aining in ec ea ional unne s. Eu opean Jou nal o Spo Science, Ea ly online. h ps://doi.o g/10.1002/ejsc.12115 2024 Recei ed: 31 Janua y 2024 - Re ised: 13 Ma ch 2024 - Accep ed: 8 Ap il 2024 DOI: 10.1002/ejsc.12115 ORIGINAL PAPER Moni o ing a igue s a e wi h hea a e‐based and subjec i e me hods du ing in ensi ied aining in ec ea ional unne s Olli‐Pekka Nuu ila 1,2 |A ja Uusi alo 3,4 |Veli‐Pekka Kokkonen 1 | Nilushika Wee a a hna 1 |Heikki Ky öläinen 1 1 Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Jy äskylä, Finland 2 The UKK Ins i u e o Heal h P omo ion Resea ch, Tampe e, Finland 3 Depa men o Spo s and Exe cise Medicine, Clinicum, Uni e si y o Helsinki, Helsinki, Finland 4 Helsinki Clinic o Spo s and Exe cise Medicine, Founda ion o Spo s and Exe cise Medicine, Helsinki, Finland Co espondence Olli‐Pekka Nuu ila, Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, P.O. Box 35 (VIV), Jy äskylä 40014, Finland. Email: [email p o ec ed] Funding in o ma ion Pola Elec o Oy; The Finnish Spo s Resea ch Founda ion; The Founda ion o Spo s Ins i u e Abs ac The pu pose o his s udy was i s ly o examine he sensi i i y o hea a e (HR)‐ based and subjec i e moni o ing ma ke s o in ensi ied endu ance aining; and secondly, o in es iga e he alidi y o hese ma ke s o dis inguish indi iduals in di e en a igue s a es. A o al o 24 ec ea ional unne s pe o med a 3‐week baseline pe iod, a 2‐week o e load pe iod, and a 1‐week eco e y pe iod. Pe o - mance was assessed be o e and a e each pe iod wi h a 3000m unning es . Re- co e y was moni o ed wi h daily o hos a ic es s, noc u nal HR eco dings, ques ionnai es, and exe cise da a. The pa icipan s we e di ided in o subg oups (o e eached/OR, n=8; esponde s/RESP, n=12) based on he changes in pe - o mance and subjec i e eco e y. The esponses o he second week o he o e load pe iod we e compa ed be ween he subg oups. RESP imp o ed hei baseline 3000 m ime (p<0.001) a e he o e load pe iod (−2.5 �1.0%), and he change di e ed (p<0.001) om OR (0.6 �1.2%). The changes in noc u nal HR (OR 3.2 �3.1%; RESP −2.8 �3.7%, p=0.002) and HR a iabili y (OR −0.7 �1.8%; RESP 2.1 �1.6%, p=0.011) di e ed be ween he subg oups. In addi ion, he dec ease in subjec i e eadiness o ain (p=0.009) and inc ease in so eness o he legs (p=0.04) we e g ea e in OR compa ed o RESP. Noc u nal HR, eadiness o ain, and exe cise‐de i ed HR‐ unning powe index had ≥85% posi i e and nega i e p edic i e alues in he disc imina ion be ween OR and RESP indi iduals. In conclusion, exe cise ole ance can a y subs an ially in ec ea ional unne s. The esul s suppo ed he use ulness o noc u nal HR and subjec i e eco e y assess- men s in ecognizing a igue s a es. KEYWORDS a igue, o e aining, pe o mance, physiology, eco e y This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and ep oduc ion in any medium, p o- ided he o iginal wo k is p ope ly ci ed. © 2024 The Au ho s. Eu opean Jou nal o Spo Science published by Wiley‐VCH GmbH on behal o Eu opean College o Spo Science. Eu J Spo Sci. 2024;1–13. wileyonlinelib a y.com/jou nal/ejsc - 1 Highligh s �Tole ance o simila ly inc eased aining load a ies in ec ea ional unne s. While some indi iduals espond posi i ely, o o he s he same load may be excessi e. �Subjec i e ma ke s, such as eadiness o ain and so eness o he legs, we e he mos sensi i e ma ke s o espond o inc eased aining load, and he magni ude o hese e- sponses was g ea e in indi iduals wi h suspec ed o e eaching. �Res ing o exe cise hea a e (HR) did no espond as sys ema ically as he subjec i e ma ke s o he aining load i sel . Howe e , he changes in noc u nal HR, noc u nal HRV, and HR‐ unning powe index s a ed o di e be ween he RESP and OR subg oups a an ea lie s age o he o e load pe iod suppo ing hei use ulness as indica o s o he a igue s a e. 1 | INTRODUCTION Classic ma hema ical aining models assume ha once an indi idual is exposed o a ce ain exe cise dose, i esul s in posi i e esponses in i ness, which can be ealized a e he alle ia ion o a igue (Taha e al., 2003). The s a e o a igue is ypically conside ed as a con- inuum, and he le el o a igue is de ined based on changes in pe - o mance and he ime needed o eco e (Bellinge , 2020; Meeusen e al., 2013). Fo diagnos ic pu poses, pe cei ed eco e y o mood is o en e alua ed along wi h he pe o mance (Aub y e al., 2015; Bellinge , 2020; Coa es, Hammond, & Bu , 2018). A s a e in which he pe cei ed a igue migh be inc eased abo e no mal, bu pe o - mance is unchanged o imp o ed is called acu e a igue, and i occu s basically a e each aining session bu ades wi hin day(s). I incomple e eco e y accumula es long enough, he nex s age o a- igue is ega ded as o e eaching (OR), which in he case o unc- ional OR (FOR) akes om a ew days o a ew weeks o eco e , bu du ing non unc ional OR (NFOR) he eco e y pe iod is subs an ially longe (i.e., weeks‐ o‐mon hs). (Meeusen e al., 2013). Al hough some s udies ha e ound signi ican imp o emen s a - e he eco e y pe iod in FOR a hle es (Bellenge e al., 2016; Le Meu e al., 2013), he bene i s o p esc ibed pe iods o FOR ha e been ques ioned when compa ed o aining ha induces b ie pe- iods o acu e a igue only (Bellinge , 2020; Bellinge e al., 2020). La ely, FOR has also been connec ed o se e al un a o able physio- logical esponses, such as inc eased es ing a e ial s i ness (Coa es, Milla , & Bu , 2018), mi ochond ial unc ional impai men (Flockha e al., 2022), dec eased glucose ole ance (Flockha e al., 2022), and highe p e alence o uppe espi a o y ac in ec ions (Hausswi h e al., 2014). The e o e, ma ke s ha a e sensi i e and speci ic o he p edic ion o OR could be use ul by suppo ing he p esc ip ion o sus ainable aining p og ams. The e olu ion o wea able echnology has p o ided a lo o al- e na i es o he moni o ing o aining and eco e y. Among endu ance aining, he mos ypical echnologies a e ela ed o hea a e (HR) eco dings a es o du ing exe cise. Al hough many s udies suppo he use ulness o HR‐based ma ke s (Buchhei , 2014; Düking e al., 2021; Nuu ila e al., 2022a), esea ch e idence is somewha con lic ing ega ding hei accu acy in he iden i ica ion o a igue s a es. Acco ding o e iews by Bosque e al. (2008) and Roe e e al. (2021), HR measu emen s du ing es o exe cise would no appea o be su icien in he de ec ion o FOR, indi idually. In line wi h ha , Bellenge e al. (2016) ha e also sugges ed ha e- sponses in he HR‐based ma ke s should be con ex ualized wi h pe cei ed eco e y as simila esponses (dec ease in HR and inc ease in HRV) can be associa ed wi h posi i e aining adap a ions and he s a e o FOR. In gene al, subjec i e ma ke s ha e been ound o be mo e sensi i e han objec i e measu es in esponding o changes in aining load (Saw e al., 2016). Howe e , i is c i ical o acknowledge ha no jus he sensi i i y o a ce ain (ex e nal) aining load is impo an . E en mo e ele an would be he speci ici y o he ma ke o dis inguish excessi e load om sus ainable load o an indi idual, as he homeos a ic s ess induced by a ce ain aining load and abili y o eco e is highly a iable be ween indi iduals (Mann e al., 2014). The con lic ing esul s ega ding he abili y o HR‐based ma ke s o iden i y a igue s a es could, a leas pa ly, ela e o he me h- odology. Fo example, s udies examining es ing HR/HRV ha e mos ly ocused on mo ning eco dings (Roe e e al., 2021), while he use ulness o day‐ o‐day noc u nal eco dings emains a he un- known. Mo eo e , he assessmen o a eliable baseline is o impo - ance in HR‐based ma ke s, and ins ead o single isola ed esul s, longe a e aging pe iods migh be necessa y (Man esa‐Rocamo a e al., 2021). Some s udies ha e induced so called “pa asympa he ic hype ac i i y” a he s a e o FOR (Bellenge e al., 2016; Le Meu e al., 2013), meaning dec eased HR and inc eased HRV along wi h impai ed pe o mance. Howe e , i is inconclusi e how condi ional he phenomenon is o me hodologies ( eco ding condi ion), aining execu ion ( equency, in ensi y, and olume), and backg ound o he pa icipan s (well‐ ained s. ec ea ional). Fu he mo e, as calo ic es ic ion (Ma hisen e al., 2020) and ela i e ene gy de iciency (S ein e al., 2012) ha e been associa ed wi h inc eased pa a- sympa he ic ac i i y, low ene gy a ailabili y could explain some o he con adic o ies obse ed du ing pe iods o excessi ely high aining loads. The pu pose o his s udy was o examine he sensi i i y o HR‐ based and subjec i e moni o ing ma ke s o in ensi ied endu ance aining. Fu he mo e, he capabili y o hese ma ke s o di e en ia e 2 - NUUTTILA ET AL. 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License be ween indi iduals o di e se a igue s a es and he op imal cu ‐o alues o classi y indi iduals co ec ly we e in es iga ed. 2 | MATERIALS AND METHODS 2.1 | Pa icipan s A o al o 32 (18 males and 14 emales) ec ea ional unne s we e ec ui ed o pa icipa e in he s udy. The pa icipan s we e heal hy, aged 20–45 (men) o 20–50 (women), and ained egula ly by unning. The heal h s a us o all indi iduals willing o pa icipa e was sc eened ia a ques ionnai e o exclude any diseases o egula medica ions ha could ha e a ec ed he pa icipa ion. In addi ion, hei es ing elec oca diog aphy was eco ded and app o ed by a physician be o e inal accep ance. F om he 32 pa icipan s ha we e sc eened, he e we e 8 d opou s. The easons o he d opou s we e challenges wi h he ime able (n=2), heal h issues un ela ed o he s udy (n=2), and inju ies/so eness o he legs (n=4). The baseline cha ac e is ics o he pa icipan s who pe o med he whole s udy pe iod and a ended all he es s a e p esen ed in Table 1. All he pa icipan s ga e hei w i en consen o pa icipa e, and he s udy p o ocol was app o ed by he e hics commi ee o he Uni e si y o Jy äskylä. 2.2 | S udy p o ocol The s udy consis ed o h ee dis inc phases: a 3‐week baseline aining pe iod (BL), a 2‐week o e load pe iod (OL), and a 1‐week eco e y pe iod (REC) (Figu e 1). Each pe iod was ollowed by a es day (T1–T3) du ing which maximal endu ance pe o mance was assessed wi h a 3000 m unning es . In addi ion, amilia iza ion es s (T0) and a maximal inc emen al eadmill es we e pe o med be o e he BL. T aining and eco e y we e moni o ed h oughou he s udy pe iod by mo ning o hos a ic es s and nigh HR and HRV e- co dings, subjec i e eco e y ques ionnai es, exe cise da a, and weekly con ol unning es s (C1–C6). 2.3 | T aining p o ocol The aining p og am was designed based on p e ious aining in- e en ions aiming o induce a momen a y o e load (Bellenge e al., 2016; Bou dillon e al., 2018; Dupuy e al., 2013; Le Meu e al., 2013). Fu he mo e, he aim was o induce o e eaching in some indi iduals bu also allow some indi iduals o espond posi- i ely despi e compa able inc ease in hei aining load. Du ing BL, he pa icipan s exe cised a hei egula equency and olume (4–6 sessions pe week). The in ensi y o he aining was limi ed o low in ensi y (HR < i s lac a e h eshold), excep o he weekly con ol exe cise which consis ed o a sel ‐paced wa m‐up and 6x3‐min TABLE 1Baseline cha ac e is ics o he pa icipan s. Females (n=10) Males (n=14) All (n=24) Age (y s.) 39 �8 39 �5 39 �6 Heigh (cm) 165 �8 179 �7 173 �10 Body mass (kg) 62.6 �8.5 80.0 �12.7 72.8 �14.0 Fa (%) 24.7 �5.2 14.7 �3.8 18.8 �6.7 VO 2max (ml/kg/min) 43.1 �3.4 51.2 �8.3 47.8 �7.8 FIGURE 1 S udy design. T0–T3 e e s o es days, when he 3000 m unning es and eac i i y jump es we e pe o med. C1–C6 e e s o con ol unning es s which consis ed o a submaximal unning es , coun e mo emen jump es , and 6x3‐min high‐in ensi y in e al aining session. Du ing he baseline pe iod, aining load o he 3 d week was sligh ly dec eased (−20%) o ensu e su icien eco e y be o e he o e load pe iod. EUROPEAN JOURNAL OF SPORT SCIENCE - 3 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License maximum sus ainable e o in e als. The in ensi y es ic ion was in ended o ensu e su icien aining eadiness o OL. Du ing he OL, he aining load was inc eased ia bo h he in ensi y and he olume o aining. The planned inc ease in he aining load was 80% compa ed o BL as de e mined by he o - mula o Lucia e al. (2003) whe e he minu es spen a di e en in ensi y zones (zone 1 =1, Zone 2 =2, and Zone 3 =3) ha e hei own weigh coe icien s. An inc ease o aining olume was s an- da dized o 45%. The weekly aining p og am du ing he OL con- sis ed o a con ol session, 3–4 mode a e‐in ensi y sessions (HR be ween lac a e h esholds), and 1–2 low‐in ensi y sessions. The mode a e‐in ensi y sessions we e 6‐and 12‐min in e als wi h a 3‐ min ac i e eco e y o con inuous >30 min sessions. All he mode a e‐in ensi y sessions included a 15–20‐min wa m‐up and cool‐down. The aining equency and he o al olume o each aining session we e de ined acco ding o he aining du ing he BL and he abo e‐men ioned c i e ia. A e he OL, he aining load as Lucia's TRIMP was dec eased du ing he REC by 40% compa ed o BL. Excep o he con ol session, he aining was pe o med as low‐in ensi y aining. 2.4 | Pe o mance es s An inc emen al eadmill es was pe o med be o e he s a o BL o de e mine indi idual HR zones. Du ing he same isi , he pa ici- pan 's a pe cen age was es ima ed wi h skin old measu emen s (Du nin e al., 1967). The eadmill es s a ed a 7 km/h ( emales) o 8 km/h (males), a e which he eadmill speed was inc eased by 1 km/h e e y 3 min, and he es con inued un il oli ional exhaus- ion. The incline was kep cons an a 0.5°. The eadmill was s opped be ween each s age o d awing blood samples om he inge ip o lac a e analyses (Biosen S_line Lab þlac a e analyze , EKF Diag- nos ic). The HR (Pola H10, Pola Elec o Oy) and espi a o y gases (Jaege Vyn us CPX, Ca eFusion Ge many 234 GmbH) we e also measu ed con inuously du ing he es . The maximal oxygen up ake (VO 2max ) was de ined as he highes 60‐s a e age o oxygen up ake and he maximum HR as he highes obse ed alue du ing he es . The lac a e h esholds (LT1 and LT2) we e de ined acco ding o he c i e ia used in p e ious s udies (Nuu ila e al., 2022a,2022b; Ves e inen e al., 2017). T0–T3 es s ( eac i i y jumps þ3000 m) ook place always a he same ime o he day wi hin‐indi idual (�2 h), and hey we e always p eceded by a es day. Pa icipan s we e asked o ollow simila nu i ional habi s on all es days. Since i is known ha low ene gy a ailabili y can pa ly con ibu e o o e eaching (Ke unen e al., 2021; Kuikman e al., 2022), he pa icipan s we e ad ised o pay a en ion o su icien ene gy in ake du ing OL. Mo eo e , hey ecei ed a calcula ion o he app oxima e inc ease in he daily ene gy expendi u e ha was caused by he inc eased aining load. Body mass ha was measu ed on a scale be o e each es was used as an indi ec measu e o es ima e i he e had been indi iduals wi h sig- ni ican ene gy de iciency. Reac i i y jump (RJ powe ) es s we e pe o med a e a 3‐min wa m up and wo submaximal se s o ~10 jumps. In he es , he pa icipan s pe o med 10 epea ed jumps explosi ely wi h minimal con ac ime. Hands we e held on hips and minimal bending on he knees was allowed. Con ac imes and ligh imes we e analyzed om a ideo ha was eco ded wi h slow‐mo ion came a (240 ps) om he on and analyzed wi h MyJump Lab (Haynes e al., 2019). The powe o each jump was calcula ed based on he o mula o Bosco e al. (1983). The a e age o i e bes jumps was used in he analyses. The in e and in aday eliabili y o he 10/5 epea ed jump es p o ocol has been epo ed in he a hle ic popula ion (Sou hey e al., 2023). 3000 m unning es s we e conduc ed in small g oups (max. Six pe sons) in a 200‐m indoo ack (n=18) o in a 400‐m ou doo ack (n=6) a e he eac i i y jump es . The ou doo ack was used o some pa icipan s (in all es s) due o he summe lockdown o he indoo ack ha was no known when he ime able o he da a collec ion was designed. A s anda dized 15‐min low‐in ensi y wa m‐ up including 3 �20–30‐s accele a ions o he a ge speed was al- ways pe o med be o e he es . Ve bal encou agemen and spli imes we e gi en o all pa icipan s. The unning ime, a e age and peak HR, and blood lac a e concen a ion (pos 2 min) we e analyzed o each es . The 3000 m unning es was chosen as an indica o o endu ance pe o mance due o i s easibili y, p e iously epo ed as low coe icien o a ia ion (CV) (1.41%) in ec ea ional unne s (Nuu ila e al., 2022b) and s ong associa ions wi h inc emen al eadmill es a iables (Nuu ila e al., 2023). 2.5 | T aining and eco e y moni o ing The pa icipan s used an HR moni o (Van age V2, Pola Elec o Oy) and HR s ap (H10) in all aining sessions. The ime in HR zones (1 <LT1, 2 =LT1‐LT2, and 3 >LT2) and he dis ance co e ed we e analyzed o all he sessions. In addi ion, he a e age HR and unning powe o whole sessions (con inuous low‐in ensi y) o wa m‐up (mode a e‐in ensi y) we e analyzed o calcula e a simila a iable as he HR‐ unning speed index (Ves e inen e al., 2014) by eplacing he unning speed wi h unning powe in he o mula. The HR‐ unning powe index basically demons a es he di e ence be- ween heo e ical and measu ed unning powe a a ce ain HR based on he indi idual's s anding HR (BL a e age), maximum HR, and maximum speed o he inc emen al eadmill es con e ed o unning powe in he Pola Flow so wa e. The unning powe was used ins ead o unning speed o allow a ai e compa ison be ween possibly di e ing e ains. An example o HR‐ unning powe index calcula ion is demons a ed in he Suppo ing In o ma ion S1. Noc u nal HR eco dings we e pe o med wi h alida ed (Nuu ila e al., 2021) pho ople hysmog aphy‐based w is measu emen (Pola Van age V2). The esul s (HR, LnRMSSD) we e analyzed sepa a ely o he ull nigh and a 4‐h pe iod which s a ed 30 min a e he de ec ed sleep onse (Nuu ila e al., 2022c) ha was based on an au oma ic algo i hm (Pesonen e al., 2018). The esul s we e i s 4 - NUUTTILA ET AL. 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License synced om he wa ch o Pola Flow so wa e, a e which consec- u i e 5‐min a e ages o he noc u nal eco ding da a we e au o- ma ically ans o med o Coach4P o so wa e (Coach4P o Oy, Espoo, Finland). In he o hos a ic es , he pa icipan s collec ed RR‐in e al da a o 2 min in he supine and s anding posi ions. Be o e he measu e- men , he pa icipan s we e ins uc ed o emp y hei u ina y bladde and a e ha go back o bed and wai o app oxima ely 1 min be o e s a ing he eco ding. The da a was collec ed by using an ECG‐based HR‐senso (Pola H10) and he “o hos a ic es ” ea u e o he wa ch (Pola Van age V2). The esul s we e analyzed o mean HR and LnRMSSD du ing he second minu e o he supine and s anding posi ions. The con ol unning es consis ed o a a ing o a pe cei ed exe ion (RPE)‐based submaximal unning es , coun e mo emen jumps (CMJs), and a 6x3‐min maximal sus ainable e o in e al exe cise. The es was ins uc ed o be pe o med once a week in he same o simila en i onmen and a he same ime o day (�2 h) wi hin an indi idual. The RPE‐based es was de eloped om he p o ocols o Sangan e al. (2021) and Ves e inen e al. (2017). I in ol ed wo 6‐min s ages and one 3‐min s age wi h in ensi ies de ined on he RPE scale as 9 ( e y ligh ), 13 (somewha ha d), and 17 ( e y ha d). The a e age unning speed and he a e age HR we e calcula ed sepa a ely o each s age (excluding he i s minu e o each s age). The CMJ es (Pola Leg Reco e y Tes ) consis ing o h ee maximal a emp s was pe o med be o e in e al exe cise. The jump heigh was es ima ed au oma ically ia he ine ial measu e- men uni in he wa ch (Pola Van age V2). The alidi y and eliabili y o he me hod ha e been epo ed by G ube e al. (2022). A e he CMJ es , he pa icipan s pe o med a 6x3‐min/2‐min eco e y in- e al session a he maximum sus ainable e o . The a e age unning speed and he a e age HR we e de e mined as he a e age o all in e als. The a e age speed o he in e als has p e iously co ela ed s ongly wi h he 3000 m unning pe o mance and i s change (Nuu ila e al., 2023). Subjec i e eco e y was assessed each mo ning a e he o ho- s a ic es ia he Coach4P o applica ion. The a iables es ima ed on a 1–5 scale we e as ollows: sleep quali y, gene al a igue, s ess le el, eadiness o ain, so eness o he legs, and a igue o he leg muscles. The scale logic was adap ed om Hoope e al. (1995): Numbe h ee indica ed no mal pe cep ion, while inc ease indica ed sligh ly o much wo se pe cep ion and dec ease indica ed sligh ly o much be e pe cep ion. The pa icipan s we e ad ised o de ine no mal he same way i was es ima ed a he beginning o he s udy and no o upda e, o example, possibly inc easing a igue du ing he s udy pe iod as he “new no mal”. In addi ion, aining‐speci ic ques ions we e asked in he same ques ionnai e consis ing o ses- sion RPE on a 0–10 scale (Fos e e al., 1996) and exe cise pe cep ion (asking on a 1–5 scale) i p e ious session el as expec ed o be e o wo se han expec ed. To ease he in e p e a ion o he esul s, hey we e escaled o he analysis as ollows: No mal was de ined as 0 in all he ques ions, dec ease was de ined as a nega i e numbe down o −2, and inc ease as a posi i e numbe up o 2 ega dless o whe he he change should be conside ed as nega i e o posi i e in pe cep ion (e.g., sligh ly inc eased so eness o he legs =1 s. sligh ly imp o ed eadiness o ain =1). 2.6 | Classi ica ion o a igue s a e The classi ica ion o he a igue s a e was based on he changes in pe o mance and subjec i e eco e y. Typically, he dec emen in pe o mance ollowed by a ebound, a leas , o he baseline le el is used as a c i e ion o FOR (Bellinge , 2020; Meeusen e al., 2013). Howe e , as Coa es, Hammond, & Bu , 2018 ha e a gued, in ec- ea ionally ained indi iduals his c i e ion may no be sui able du ing o e load pe iods due o la ge inc eases in pe o mance, which can “mask” he subsequen pe o mance dec emen s. The e o e, he a igue s a e was diagnosed based on he o mula p oposed by Coa es, Hammond, & Bu , 2018 bu modi ied o cu en es p o- ocol as ollows: 3000 m ime (Δ%) [(T1–T2) þ(T3–T2)]. The o mula combines he dec ease in unning pe o mance expec ed om T1 o T2 wi h he supe ‐compensa ion expec ed a T3. I he change de i ed om he o mula was g ea e han he smalles wo hwhile change (0.3 �CV o T0–T1) (Hopkins, 2004) and he subjec i e eadiness o ain was dec eased compa ed o BL ( en Haa e al., 2017), he indi idual was de ined as o e eached (OR). A o al o eigh pa icipan s (4 emales and 4 males) ul illed his c i e ion. To compa e he OR subg oup wi h indi iduals who comple ed simila inc emen in aining load du ing OL bu imp o ed hei pe o mance om T1 mo e han he smalles wo hwhile change a T2 and T3, a subg oup o esponde s (RESP) was o med. A o al o 12 pa icipan s (3 emales and 9 males) ul illed his c i e ion lea ing ou pa ici- pan s who could no be classi ied as OR o RESP ou o he subg oup analysis. One o hese pa icipan s ul illed he pe o mance c i e ion o OR, bu he subjec i e eadiness o ain imp o ed signi ican ly om BL o OL. Two pa icipan s ul illed pe o mance c i e ion o RESP a T2 bu no a T3 due o impai ed pe o mance. One pa icipan did no imp o e pe o mance mo e han SWC a T2 bu did no ha e supe compensa ion a T3 ei he . Ins ead o classi ying unclea pa icipan s on case‐by‐case basis as OR o RESP, i was conside ed mo e app op ia e o s ick o objec i e, ep oducible c i e ia ha would allow mo e accu a e compa ison be ween in- di iduals who in p ac ical e ms had meaning ully di e ing a igue s a es. 2.7 | S a is ical analyses The esul s a e p esen ed as mean �s anda d de ia ion. The no mali y o he da a was analyzed wi h Shapi o–Wilk es . The e - ec s o he OL and REC pe iods we e analyzed by epea ed mea- su es ANOVA wi h Bon e oni pos hoc es o i he a iables we e no no mally dis ibu ed ( aining cha ac e is ics) by F iedman nonpa ame ic es s. Fo he pe o mance es s, compa isons we e pe o med be ween he T1–T3 ime poin s, and o he moni o ing EUROPEAN JOURNAL OF SPORT SCIENCE - 5 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License a iables, he weekly a e ages du ing he OL (OL1 and OL2) and REC pe iods we e compa ed o he a e age o BL. Weekly a e ages ha e also been used in p e ious o e load s udies (Bellenge e al., 2016; Bou dillon e al., 2018; Le Meu e al., 2013), and hey been sug- ges ed o be me hodologically supe io compa ed o single‐day e- sul s in he de ec ion o FOR (Man esa‐Rocamo a e al., 2021). Be ween‐subg oup (OR s. RESP) di e ences in he changes om T1 o BL we e examined wi h simple con as s o in nonpa ame ic es s wi h a Mann–Whi ney U‐ es . In addi ion, a ecei e ope a ing cha ac e is ic (ROC) analysis (Hajian‐Tilaki, 2013) was pe o med o examine how well each weekly measu ed moni o ing ma ke was able o classi y he pa icipan nega i ely o posi i ely as o e - eached. Fo he ma ke s ha had signi ican (p<0.05) a ea unde he ROC cu e (AUC), an op imum cu ‐o alue was analyzed based on he maximum Kolmogo o –Smi no me ic. Finally, using his cu ‐ o alue, posi i e and nega i e p edic i e alues we e analyzed. In he esul s o es ing HR and HRV, one pa icipan (in OR subg oup) was excluded om he analyses due o abno mally high HRV ha p e en ed ob aining da a om he wa ch eco dings du ing he s udy pe iod. S a is ical analyses we e pe o med wi h IBM SPSS S a is ics .28 p og am (SPSS Inc). 3 | RESULTS 3.1 | T aining The mean weekly aining cha ac e is ics o he o al g oup o pa - icipan s and subg oups du ing each pe iod a e p esen ed in Table 2. The weekly aining olume inc eased om BL du ing OL 44 �5% (p<0.001). In u n, Lucia's TRIMP inc eased by 73 �11% (p<0.001) and session RPE by 122 �44% (p<0.001). No di e ences we e ound be ween he subg oups ega ding any aining cha ac e is ics du ing he OL (p>0.157). 3.2 | Running and jumping pe o mance T1 3000‐m ime (p=0.610) o RJ powe (p=0.910) did no di e be ween he subg oups. The 3000‐m ime dec eased signi ican ly in he o al g oup o pa icipan s om T1 o T2 (−1.2 �1.7%, p=0.006) and T1 o T3 (−1.7 �1.2%, p=0.002). The changes om T1 o T2 (OR 0.6 �1.2%; RESP −2.5 �1.0%) and T1 o T3 (OR ‐0.8 �1.6%; RESP −3.1 �0.8%) di e ed be ween he subg oups (p<0.001), and he unning ime emained unchanged in OR a T1–T3. The blood lac a e dec eased om T1 o T2 only in he OR g oup (p=0.03), and he change di e ed om he RESP g oup (OR −2.0 �2.4 mmol/L; RESP 0.3 �1.6 mmol/L, p=0.02). In HRa g o HRpeak, he e we e no signi ican changes in nei he o he subg oups, bu endencies owa d dec emen om T1 o T2 in HRa g (p=0.07) and HRpeak (p=0.06) we e ound in he OR. The RJ powe did no change in ime, and no di e ence be ween he subg oups was obse ed. The abso- lu e unning and jumping pe o mance es esul s a e p esen ed in he Suppo ing In o ma ion S2, and indi idual esponses in 3000‐m wi hin he subg oups a e p esen ed in Figu e 2. The body mass emained unchanged h ough he s udy pe iod in he o al g oup o pa icipan s and in bo h subg oups. 3.3 | Hea a e and hea a e a iabili y in he o hos a ic es and du ing sleep The s anding HR dec eased om BL o OL2 (p=0.02), while he noc u nal 4‐h HR dec eased only om BL o REC (p=0.02) in he o al g oup o pa icipan s. In addi ion, he noc u nal 4‐h HRV inc eased om BL o OL2 (p=0.04) and ended o emain inc eased a REC (p=0.054). Subg oup analyses a e p esen ed in Figu e 3. The e was a dec ease om BL o OL2 in he s anding HR o RESP (p=0.005), and he change ended o be di e en compa ed o OR (p=0.054). In addi ion, he noc u nal 4‐h HRV inc eased in RESP om BL o OL2 TABLE 2Weekly mean aining cha ac e is ics du ing he baseline (BL), o e load (OL), and eco e y (REC) pe iods. ALL (n=24) RESP (n=12) OR (n=8) Volume (h) BL 4.5 �1.0 4.4 �0.9 4.9 �1.3 OL 6.5 �1.3*** 6.4 �1.2*** 7.0 �1.7*** REC 2.7 �0.8** 2.6 �0.8*** 2.9 �0.9*** F equency (sessions/wk) BL 4.3 �0.7 4.3 �0.7 4.7 �0.8 OL 5.4 �0.7*** 5.4 �0.6*** 5.6 �0.7*** REC 3.4 �0.6** 3.3 �0.5*** 3.5 �0.7*** Dis ance (km) BL 43 �14 41 �12 50 �17 OL 65 �20*** 63 �18*** 73 �23*** REC 26 �8** 24 �6*** 29 �12*** HRz1/z2/z3 (%) BL 89/8/4 88/8/3 91/6/3 OL 68/26/6*** 67/27/5*** 76/21/4*** REC 85/11/5* 83/12/4 88/8/4 Lucia's TRIMP BL 310 �58 304 �50 328 �77 OL 533 �96*** 528 �95*** 535 �119*** REC 193 �51** 192 �61** 198 �49*** Session RPE BL 1034 �206 1066 �185 1075 �2227 OL 2297 �623*** 2356 �584*** 2301 �623*** REC 711 �362*** 787 �484*** 666 �151*** No e: RESP = esponde s and OR =indi iduals wi h suspec ed o e eaching. ***p<0.001, **p<0.01, and *p<0.05 compa ed o BL. 6 - NUUTTILA ET AL. 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License (p=0.01), and he change di e ed compa ed o OR (p=0.011). The noc u nal 4‐h HR did no change wi hin he g oups, bu he be ween‐ g oup change om BL o OL2 was di e en (p=0.002). 3.4 | Changes in submaximal unning pe o mance, con ol unning es , and CMJ The HR‐ unning powe index inc eased in he o al g oup o pa ic- ipan s om BL o OL2 (p<0.001) and REC (p=0.001). In he subg oup analyses, HR‐ unning powe index inc eased om BL o OL2 and REC only in RESP (p<0.001), and he change di e ed om OR a OL2 (p=0.052) and REC (p=0.009). The HR du ing he con ol unning es dec eased om BL o OL2 a RPE 13 (p=0.001), RPE 17 (p<0.001), and du ing 6x3‐min in e als (p<0.001) in he o al g oup o pa icipan s. The speed emained unchanged a all RPE le els and du ing he 6x3‐min in- e als ac oss he s udy pe iod. The e we e no signi ican di e ences be ween he subg oups in any o he con ol es pa ame e s (Figu e 4). FIGURE 2 Indi idual esponses wi hin he subg oups. Whi e plo s e e o indi idual “ esponde s” (RESP) and g ay plo s “o e eached” (OR) indi iduals. FIGURE 3 Hea a e (HR) and HR a iabili y (LnRMSSD) in he o hos a ic es (A, B) and du ing sleep (C, D) in he esponde s (RESP) and indi iduals wi h suspec ed o e eaching (OR). BL =baseline aining pe iod; OL =o e load aining pe iod; REC = eco e y pe iod. Dashed ci cle indica es signi ican be ween‐g oup di e ence in he change om BL. **p<0.01 and *p<0.05 compa ed o he baseline (BL) alues. EUROPEAN JOURNAL OF SPORT SCIENCE - 7 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License The CMJ ha was assessed wi hin he con ol sessions dec eased om BL o OL1 (−5.9 �7.9%, p<0.01) and OL2 (−7.7 �8.7%, p=0.003) in he o al g oup o pa icipan s. The dec ease om BL o OL2 was no signi ican in OR (−9.7 �11.0%, p=0.105) o RESP (−7.4 �8.5%, p=0.082). 3.5 | Subjec i e eco e y Subjec i e eadiness o ain dec eased om BL o OL2 (p<0.001), while so eness o he legs (p=0.04) and a igue o he leg muscles inc eased (p<0.001) a OL2 compa ed o BL in he o al g oup o pa icipan s. All subjec i e a iables e u ned o baseline a REC and we e no di e en compa ed o BL. The changes in subjec i e e- co e y wi hin he subg oups a e p esen ed in Figu e 5. Al hough subjec i e eadiness o ain dec eased om BL o OL2 in OR (p<0.001) and RESP (p=0.004), he changes we e g ea e in OR compa ed o RESP (p=0.009). Simila ly, he change in so eness o he legs om BL o OL2 di e ed be ween he subg oups (p=0.04). The exe cise pe cep ion dec eased om BL o OL2 only in he OR g oup (p=0.009) and he change di e ed om RESP (p=0.02). Subjec i e sleep, a igue, and s ess emained una ec ed compa ed o BL du ing OL1 and OL2 in he o al g oup and subg oups o he pa icipan s. 3.6 | Disc imina ion o OR and RESP indi iduals The a iables wi h he mos signi ican esul s in he ROC analyses a e p esen ed in Figu e 6, and hei 7‐day olling a e age esponses ac oss he OL a e p o ided in he Suppo ing In o ma ion S3. In addi ion o he a iables p esen ed in he igu es, he AUC o supine HR (0.821, cu ‐o 0.80%, p=0.006), noc u nal ull HR (0.845, cu ‐o 0.68%, p=0.001), and noc u nal ull LnRMSSD (0.738, cu ‐o 2.23%, p=0.038) we e signi ican . When OL‐induced esponses o moni o ing a iables we e compa ed be ween he subg oups, changes in HR‐ unning powe index s a ed o di e a e 6 days, in noc u nal 4‐h HR a e 9 days, and in noc u nal 4‐h HRV a e 10 days o OL. Among he subjec i e ma ke s, changes in eadiness o ain s a ed o di e on he 11 h day, while so eness o he legs and exe cise pe cep ion di e ed only on he 14 h day o OL. The mos alid a iables o disc imina e be- ween OR and RESP indi iduals a he end o he OL we e noc u nal HR, eadiness o ain, and HR‐ unning powe index, all ha ing ≥85% FIGURE 4 Speed and ela i e hea a e (HR) du ing he a ing o pe cei ed exe ion (RPE)‐based submaximal unning es and he ollowing 6x3‐min in e al session which was pe o med a maximal sus ainable e o . BL =baseline aining pe iod; OL =o e load aining pe iod; REC = eco e y pe iod; RESP = esponde s; and OR =indi iduals wi h suspec ed o e eaching. *p<0.05 compa ed o he baseline (BL) alues. 8 - NUUTTILA ET AL. 15367290, 0, Downloaded om h ps://onlinelib a y.wiley.com/doi/10.1002/ejsc.12115 by Uni e si y O Jy äskylä Lib a y, Wiley Online Lib a y on [02/05/2024]. See he Te ms and Condi ions (h ps://onlinelib a y.wiley.com/ e ms-and-condi ions) on Wiley Online Lib a y o ules o use; OA a icles a e go e ned by he applicable C ea i e Commons License