Incidence and treatment approach of intraocular pressure elevation after various types of local steroids for retinal diseases

Vol.:(0123456789) 1 3 G ae e's A chi e o Clinical and Expe imen al Oph halmology h ps://doi.o g/10.1007/s00417-023-06163-5 GLAUCOMA Incidence and ea men app oach o in aocula p essu e ele a ion a e  a ious ypes o local s e oids o  e inal diseases Aga aAnnaWyk o a1 · AlaaDinAbdin1· C is ianMun eanu1· U sulaLöw1· Be holdSei z1 Recei ed: 19 Feb ua y 2023 / Re ised: 13 June 2023 / Accep ed: 28 June 2023 © The Au ho (s) 2023 Abs ac Pu pose Fo he ea men o macula edema, in addi ion o he use o an i ascula endo helial g ow h ac o s, s e oids a e also used in a i eally and sub-Tenon. Side e ec s include among o he s ca a ac o ma ion and ele a ion o in aocula p essu e (IOP). The aim o his e ospec i e s udy was o elici he IOP ele a ion a e adminis a ion o a ious s e oidal medica ion, he ime o onse , and he e icacy o he adminis e ed IOP-lowe ing he apies. Me hods We included 428 eyes wi h a pos ope a i e (n = 136), diabe ic (n = 148), u ei ic macula edema (n = 61), and macula edema a e e inal ein occlusion (n = 83). These pa ien s we e ea ed wi h one o mo e di e se s e oidal agen s once o mul iple imes. These d ugs included: iamcinolone ace onide (TMC) as in a i eal injec ion (TMC IVI) o sub- Tenon (TMC ST), as well as dexame hasone (DXM) and luocinolone ace onide (FA) in a i eally. An inc ease o IOP o ≥ 25mmHg was designa ed as pa hological. A s e oid esponse in anamnesis, he ime o onse o IOP ise om he i s adminis a ion, and he he apy adminis e ed we e documen ed. Resul s O 428 eyes, 168 eyes (39.3%) had IOP ele a ion up o a mean o 29.7 (SD ± 5.6) mmHg, which occu ed a a median o 5.5mon hs. S e oids mos equen ly leading o ise o IOP included DXM (39.1% o all eyes ecei ing ha d ug), TMC IVI (47.6%), TMC ST combined wi h DXM (51.5%), DXM wi h FA (56.8%), and TMC IVI wi h DXM (57.4%). A Kaplan–Meie analysis and he Log Rank es showed a signi ican di e ence (p < 0.001). IOP ise was ea ed as ollows: 119 conse a i ely (70.8%), and 21 su gically (12.5%, cyclopho ocoagula ion 8.3%, il e ing su ge y 1.8%, in 4 he s e oidal d ug implan was emo ed 2.4%), and 28 eyes ecei ed no he apy (16.7%). Su icien IOP egula ion was achie ed in 82 eyes (68.9%) wi h opical he apy. In 37 eyes (31.1%) wi h pe sis en ly ele a ed in aocula p essu e, opical he apy had o be con inued o e he ollow-up o 20 ± 7mon hs. Conclusions IOP inc eases a e any ype o s e oid applica ion a e no a e. Resul s o ou s udy le us suspec ha especially he apy wi h in a i eal dexame hasone, ei he as a mono he apy o in combina ion wi h ano he s e oid, ends o inc ease IOP mo e han o he s e oids. Regula IOP checks a e necessa y a e each s e oid adminis a ion, wi h possible ini ia ion o long- e m conse a i e and/o su gical he apy i necessa y. Keywo ds Macula edema· S e oidal agen s· Ocula hype ension· Seconda y ocula hype ension * Aga a Anna Wyk o a Aga a.wyk o [email protected] 1 Depa men o Oph halmology, Saa land Uni e si y Medical Cen e (UKS), Hombu g/Saa , Ge many G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 In oduc ion Glaucoma is conside ed o be he second leading cause o isual impai men and blindness wo ldwide, a ec ing abou 70 million people [1]. The mos common o m o his disease, conce ning 70% o cases, is p ima y open-angle glaucoma [2]. A majo isk ac o o he disease is an ele a ed in aocula p essu e (IOP), al hough some cases p esen wi h no mal IOP [3]. In any case, eyes wi h bo h high ension and no mal en- sion glaucoma a e cha ac e ized by a p og essi e loss o e inal ne e ibe cells wi h subsequen educ ion o he isual ield [2] and educing IOP is conside ed an e ec i e ea men [4]. The apeu ic use o glucoco icoids can cause an ele a ion o IOP called s e oid-induced ocula hype ension (SIOH) and he eby s e oid-induced – seconda y – glaucoma (SIG) by ini ialing signaling cascades a ec ing exp ession o genes, which causes a highly pe sonalized pha macological esponse [5]. I was i s in 1950 when an ele a ion o IOP was documen ed a e sys emic adminis a ion o ad enoco ico ophic ho mone [6]. An inc ease o IOP a e he opical adminis a ion o co isone was no iced i s in 1954 [7]. Since hen, he phenomenon o SIOH and SIG has been s udied in ensi ely and isk ac o s, he pa hophysiology, as well as ea men we e in es iga ed. Nowa- days, s e oidal d ug implan s a e a known ea men op ion o e inal diseases, such as macula edema [8]. An inc ease in IOP unde s e oid he apy, esul ing in a seconda y glaucoma, can be obse ed in abou 30% o he popula ion. This phenomenon is called "s e oid esponse". In abou 5%, a "high- esponse" is p esen , wi h a p essu e inc ease o mo e han 15mmHg abo e he baseline p essu e [9]. I he s e oid-induced ocula hype en- sion esul ing om he s e oid he apy is o a signi ican magni- ude, and le uncon olled and un ea ed, i can lead o s e oid- induced glaucoma wi h glaucoma ous op ic neu opa hy [10]. The e a e many condi ions leading o macula edema, ou o which a e men ioned in his pape : diabe es, e inal Key messages Wha is known: Apa om ascula endo helial g ow h ac o s (an i-VEGF), s e oids a e also used in a i eally and sub-Tenon o he ea men o macula edema wi h an ele a ion o in aocula p essu e (IOP) as one he mos consequen ial side e ec s. In his s udy, 39.3% o eyes had an IOP ele a ion up o a mean o 29.7 (SD ± 5.6) mmHg, which occu ed a a median o 5.5 mon hs. Wha is new: S e oids which mos equen ly led o ise o IOP o ≥ 25 mmHg included dexame hasone (39.1% o all eyes ecei ing ha d ug), iamcinolone in a i eal (47.6%), iamcinolone sub-Tenon combined wi h dexame hasone (51.5%), dexame hasone wi h luocinolone ace onide (56.8%), and iamcinolone in a i eal wi h dexame hasone (57.4%). Rise in IOP could be ea ed in 70.8% wi h opical he apy. ein occlusion, in lamma o y eye diseases (u ei is) and s a- us pos eye su ge y. Al hough ha ing a ious pa hophysiol- ogy, hey can be all ea ed wi h s e oidal d ugs. The aim o his e ospec i e s udy was o elici he in aocula p essu e inc eases a e s e oid adminis a ion o e inal disease, he ime o onse , as well as he e i- cacy o he adminis e ed eye p essu e-lowe ing he apeu ic app oaches. Me hods S udy design andda abase The da a collec ion was pe o med e ospec i ely on 428 eyes o 349 pa ien s in he pe iod om 01.01.2016 o 31.08.2021. The s udy was pe o med in a single cen e , he Depa men o Oph halmology a he Saa land Uni e si y Medical Cen e (UKS) in Hombu g/Saa . Pa ien s we e iden i ied using he FIDUS (ou elec onic pa ien eco ds) [11] and SAP (Sys- ems, Applica ions and P oduc s in da a p ocessing – ou in e nal hospi al in o ma ion sys ems). Inclusion c i e ium o his s udy was a diagnosis o macula edema as a esul o s a us pos eye su ge y, diabe ic macula edema, e inal ein occlusion, and u ei is. Pa ien s wi h con i med p ima y open- angle glaucoma we e included. Mino s (pa ien s < 18yea s old), howe e , we e excluded om he s udy. Pe o med su - ge ies leading o seconda y macula edema included ca a ac su ge y, pa s plana i ec omy (PPV), hese wo p ocedu es combined, Desceme memb ane endo helial ke a oplas y (DMEK), and pene a ing ke a oplas y (PK). Re inal ein occlusion included bo h cen al (CRVO) and b anch e inal ein occlusion (BRVO). Macula edema was diagnosed based on undoscopy, op ical cohe ence omog aphy and luo escein G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 angiog aphy, which we e pe o med be o e he i s s e oidal d ug applica ion. Macula edema was ea ed wi h s e oidal agen s once o mul iple imes, and wi h one o mo e s e - oids. I a subjec p esen ed wi h a s e oid esponse, i.e. IOP ise, we documen ed he ime a e i s s e oid adminis a- ion and he he apy adminis e ed. An IOP ise was desc ibed as ≥ 25mmHg [12]. The he apies adminis e ed in ou s udy we e: (1) conse a i e wi h eye d ops and (2) su gical, includ- ing cyclopho ocoagula ion, il e ing su ge y and emo al o s e oidal d ug implan . A e wa ds, we documen ed whe he he chosen ea men was bene icial o he eye p essu e o no . IOP o ≤ 20mmHg was conside ed as well- egula ed and he ime a e he in oduc ion o he apy was documen ed. In case o conse a i e eye d op he apy, we also documen ed whe he a pa ien was able o s op aking he medica ion. This s udy and all in es iga ional p o ocols we e app o ed by he Saa land Uni e si y Medical Cen e as well as he E hics Commi ee o he Medical Associa ion o Saa land, Ge many (N . 123/20, da e: 16.06.2020). Ta ge igu es All pa ien s had ollow-ups o 24mon hs wi h isi s a day 1 ( i s s e oid adminis a ion), 6–8weeks, 6, 12, 18 and 24mon hs. Oph halmological examina ions A comple e oph halmological clinical e alua ion pe o med o all pa ien s included he bes co ec ed isual acui y (BCVA), in aocula p essu e (IOP) measu emen using Goldmann applana ion onome y, and an e io and pos e- io segmen examina ion wi h he sli lamp (Sli Lamp BX 900®, Haag-S ei , Köniz, Swi ze land). Macula edema wi h i s cen al macula hickness (CMT) was measu ed wi h macula op ical cohe ence omog aphy (M-OCT) (Spec a- lis OCT, Heidelbe g Enginee ing, Heidelbe g, Ge many). Fluo escein angiog aphy (FA, Heidelbe g Enginee ing, Heidelbe g, Ge many) was usually pe o med once a he ini ial p esen a ion o con i m he diagnosis. I necessa y – i was epea ed. Applica ion p ocess In his s udy, macula edema was ea ed wi h iamcinolone ace onide sub-Tenon (TMC ST) o in a i eally (TMC IVI), dexame hasone in a i eally (DXM) and luocinolone ace- onide in a i eally (FA) used in mono he apy o di e - en combina ions in ou Macula Ou pa ien Depa men (IVI cen e [13]). Physicochemical cha ac e is ics o each d ug a e summa ized in Table1. Pa ien s we e examined be o e each s e oidal d ug applica ion o check o possible con aindica ions o he injec ion, such as any kind o eye in lamma ion o o he pa hologies. We examined he BCVA, IOP, an e io and pos e io ocula segmen and educa ed he pa ien s abou he injec ion, possible isks and complica ions as well as had hem sign he consen o m. A e wa ds, he eye o be injec ed was ma ked and d opped wi h opicamide, phenyleph ine, o dila e he pupil, and polyhexanide 0.02% eye d ops, as an an isep ic. Al oge he , he eye was d opped wi h polyhexanide wice be o e and once a e an injec ion. The anamnesis and cu en he apy we e also asce ained, o ind ou abou possible changes in medica ion applied om an oph halmologis o , o example, an ele a ed IOP ha had o be ea ed. We pe o med an OCT o check he macula edema be o e he i s injec ion and hen acco ding o he ea men plan e e y one, h ee o ou appoin men s. We applied iamcinolone ace onide, dexame hasone and luocinolone ace onide in a i eally and iamcinolone ace- onide sub-Tenon unde a d op/gel anes hesia wi h p opa- acaine eye d ops and xylocaine gel o a leas 10min be o e he injec ion. 0.05ml o TMC, 700μg o DXM o 190μg o FA was en e ed empo-in e io ly wi h displacemen o he conjunc i a a 4mm om he limbus. A e wa ds, he ligh pe cep ion was checked and when posi i e – he pa ien was eleased om ou ou pa ien depa men . A e he s e - oid applica ion in a i eally, we always ecommended a Table 1 Compa ison o each s e oid d ug wi h i s ac i e subs ance, applica ion o m, ime o ac ion, maximal concen a ion (i a ailable), con- cen a ion a he end o ac ion ime (i a ailable), i s app o ed and o label use [12, 15–24] DME = diabe ic macula edema, RVO = e inal ein occlusion, PME = pos ope a i e macula edema, NI = no in o ma ion T ade name Ac i e subs ance Applica- ion Time o ac ion Maximal concen a ion (day/ alue) Concen a ion a he end o ac ion ime (day/ alue) App o ed use VOLON A T iamcinolone ace onide In a i eal injec ion 12weeks o cou- ple o mon hs 2.15 o 7.20μg/ml - VOLON A T iamcinolone ace onide Sub-Tenon injec ion 12weeks NI - OZURDEX Dexame hasone In a i eal implan 6mon hs 0.094ng/ml 0.05ng/ml DME RVO U ei is ILUVIEN Fluocinolone ace onide In a i eal implan 36mon hs 7 h day 100pg/mL DME U ei is G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 check-up appoin men in he nex wo o h ee days a ou depa men o a ex e nal doc o s´ o ices. A dose o 0.15ml o TMC was adminis e ed sub-Tenon unde he empo-in e- io conjunc i a in d op/gel anes hesia wi h p opa acaine eye d ops and xylocaine gel o a leas 10min be o e he applica ion. Fo DME and macula edema a e ein occlusion, we adminis e ed one TMC IVI as s anda d, and i no complica- ions occu ed, we pe o med in a i eal injec ion o DXM a e a leas 8weeks. DXM was usually applied e e y 4–6mon hs, while FA e e y 2–3yea s. In he absence o imp o emen a e opical and sys emic he apy o UME (u ei ic macula edema), TMC sub-Tenon ollowed by in a- i eal injec ion o DXM was adminis e ed. Pos ope a i e macula edema was p incipally ea ed wi h TMC sub-Tenon [14]. No s anda d pos ope a i e he apy was adminis e ed, especially no an ibio ics o s e oids. The apy o s e oid‑induced ocula hype ension applied in his s udy The i s -choice in aocula p essu e-lowe ing he apy applied in ou depa men we e opically adminis e ed d ugs. The ec- ommended ea men mos ly included alpha-2(α2)ad ene gic ecep o agonis s 2 o 3 imes pe day, ca bonic anhyd ase inhibi o s 2 imes pe day, be a (β-) blocke s 2 imes pe day as single medica ion, o as combina ion medica ion. I he IOP was no low enough, we p esc ibed ace azolamide 250mg in doses o 0.5, 1, 2 o 3 pe day, acco ding o needs o he pa ien , adminis e ed o ally. P os aglandin analogues we e no ecommended as hey migh igge an an e io chambe in lamma ion and macula edema, as well as i s p og ession h ough a achidonic acid and p o-in lamma o y p ocess ac i a- ion [25]. A i s , we p esc ibed a he apy wi h one medica ion (= mono he apy). I he d ug was no ole a ed i was swi ched. In con as , i he a ge IOP could no be eached – ano he d ug was added. I a pa ien s ill p esen ed wi h inc eased IOP, we decided o un a su gical, IOP lowe ing in e en ion. P ocedu es ca ied ou in ou s udy included a cyclopho o- coagula ion, abeculec omy and i ec omy o emo e a s e- oidal d ug implan and ensu e a no maliza ion o an ele a ed IOP. Anes hesia applied o pa ien s in ou s udy included a opical, e obulba , sub-Tenon and gene al anes hesia. S a is ics Da a was collec ed in a Mic oso Excel 365 sp eadshee . Desc ip i e analysis o he pa ien collec i e was al eady pos- sible using Excel. Absolu e (numbe o cases) and ela i e e- quencies (pe cen age) we e calcula ed. In addi ion, maximum, minimum, mean, and s anda d de ia ion could be calcula ed o some pa ame e s o be e desc ibe he pa ien g oups. The a i hme ic mean was always calcula ed and documen ed as he mean alue. Fo he s a is ical analysis we used IBM SPSS S a is ics 27. The Pea son Chi-Squa ed es was used o compa e ca - ego ical and o dinal a iables as well as o show a signi ican associa ion be ween g oups and ca ego ies. Fo he compa i- son o he con inuous a iable o e ime be ween numbe s o ca ego ies he Gene al Linea Model was used. Also, he pai wise compa ison be ween ime poin s was adjus ed wi h Bon e oni co ec ion. A p- alue o < 0.05 was conside ed o show a signi ican esul . Bo h eyes we e included in 79 pa ien s o 349 (22.6%), he e- o e he in e class co ela ion coe icien (ICC) [26] be ween he le and igh eye o he same pa ien was calcula ed o all pa ame e s o exclude a signi ican e ec due o co ela ed da a. Resul s S udy collec i e Demog aphic da a, gene al andeye anamnesis (Table2) The s udy included 428 eyes o 349 pa ien s, o which 188 (53.9%) we e male and 161 (46.1%) we e emale wi h an a e age age o 68.4 (± 11.9). The gene al anamnesis showed ha 251 (71.9%) subjec s p esen ed wi h hype ension, 146 (41.8%) wi h diabe es melli us wi h an a e age HbA1c o 7.5% and 159 (45.6%) ook an an icoagulan a ha ime. Ou o 428 eyes, we coun ed 236 (55.1%) igh eyes and 192 (44.9%) le ones. The p- alue o ICC o all main Table 2 Demog aphic da a, gene al and eye anamnesis o 349 pa ien s Demog aphic da a and eye anamnesis Numbe o pa ien s To al 349 Male 188 (53.9%) Female 161 (46.1%) A e age age 68.4 ± 11.9yea s Hype ension 251 (71.9%) Diabe es melli us 146 (41.8%) An icoagula ion 159 (45.6%) Numbe o eyes 428 Righ eyes 236 (55.1%) Le eyes 192 (44.9%) Eyes wi h p ima y open-angle glaucoma 65 (15.2%) Eye p essu e lowe ing eye d ops 58 (13.6%) Posi i e amily his o y o glaucoma 7 (2%) Amblyopic eyes 14 (3.3%) His o y o auma 8 (1.9%) S a us pos ca a ac su ge y 328 (76.6%) S a us pos pa s plana i ec omy 236 (31.5%) G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 ou come measu es be ween le and igh eyes was > 0.35, which indica ed a non-signi ican co ela ion be ween he wo eyes in examined pa ien s. Acco ding o he eye anam- nesis, 65 (15.2%) eyes p esen ed wi h p ima y open-angle glaucoma using only one ocula p essu e lowe ing eye d op, ou o which 7 (2%) had a posi i e amily his o y o glau- coma. 14 eyes we e amblyopic (3.3%), 8 eyes had a his- o y o auma (1.9%), 328 eyes (76.6%) had unde gone ca a ac su ge y p e iously and 135 eyes (31.5%) pa s plana i ec omy. All eyes ha expe ienced auma (n = 8, 1.9%) su e ed om a pos ope a i e macula edema la e on. Classi ica ion acco ding odiagnosis We included 136 eyes (31.8%) wi h a pos ope a i e macula edema (PME), 148 (34.6%) wi h diabe ic macula edema (DME), 83 (19.4%) wi h macula edema a e a e inal ein occlusion (RVO, bo h cen al e inal ein occlusion – CRVO, and b anch e inal ein occlusion – BRVO) and 61 (14.3%) wi h u ei ic macula edema (UME). Classi ica ion acco ding os e oids adminis e ed As a as adminis a ion o single d ug and d ug combina- ions was conce ned, he exac da a is p esen ed in Table3. Pe iod o  ollow‑up A ollow-up a 6–8weeks was done o all he pa ien s, 386 (90.2%) pa ien s we e con olled a e 6mon hs, 323 (75.5%) o hem had a ollow-up a 1yea , 256 (59.8%) o hem came in o a con ol isi a e 18mon hs, while a ull ollow-up o 24mon hs was eached by 220 pa ien s (51.4%). We conside ed he unscheduled isi s as well, as he pa ien s we e o en e e ed om a doc o because o p oblems a e s e oid adminis a ion, mos ly because o IOP ele a ion. Visual acui y (VA) The BCVA imp o ed in pa ien s wi h pos ope a i e macu- la edema wi h an a e age o 0.26 (± 0.2) decimal be o e he i s s e oid applica ion, 0.4 (± 0.26) a e 12mon hs and 0.4 (± 0.26) a e 24mon hs o ollow-up, as well as in pa ien s wi h u ei ic macula edema wi h 0.25 (± 0.13) p e- ope a i ely, 0.37 (± 0.2) a e 12mon hs and 0.42 (± 0.28) a e 24mon hs. Subjec s su e ing om diabe ic macula edema had an a e age BCVA o 0.43 (± 0.22) be o e he s e oid applica ion, 0.46 (± 0.25) a e 12mon hs and 0.38 (± 0.25) a e 24mon hs. Pa ien s a e a e inal ein occlu- sion p esen ed wi h an a e age BCVA o 0.34 (± 0.2) p e- ope a i ely, 0.37 (± 0.24) a e 12mon hs and 0.31 (± 0.26) a e 24mon hs. S a is ical analysis showed no signi ican di e ence be ween BCVA in eyes wi h an ele a ed IOP as opposed o eyes wi hou an IOP ele a ion. VA in 168 eyes wi h an ele a ed IOP was as ollows: 0.34 (± 0.21) decimal be o e he i s s e oid applica ion, 0.4 (± 0.25) a e 12mon hs and 0.35 (± 0.24) a e 24mon hs o ollow-up. Cen al macula hickness (CMT) As a as he CMT is conce ned, i was educed in all g oups unde he s e oid he apy. I amoun ed on a e age 465μm be o e he i s s e oid adminis a ion, 380μm a e 6–8weeks, 384μm a e 6mon hs, 382μm a e 12mon hs, 370μm a e 18mon hs, and 369μm a e 24mon hs. Rega ding CMT in di e en g oups be o e he s e oid appli- ca ion and a las con ol o 24mon hs, we measu ed 504μm (± 109) and 384μm (± 105) in he pos ope a i e g oup, 420μm (± 119) and 362μm (± 97) in diabe ic pa ien s, Table 3 Ele a ion o in aocula p essu e in eyes ea ed wi h mono he apy o combina ion o s e oidal agen s (TMC ST = iamcinolone sub-Tenon; FA = luocinolone ace onide; DXM = dexame hasone; TMC IVI = iamcinolone in a i eal) S e oidal d ug as mono he apy and d ug combina ion F equency (n, % o all d ug combina ions) IOP ele a ion (n, % o each d ug combina ion) Time un il IOP ise in median (mon hs) TMC ST 130 (30.4%) 30 (23.1%) 5 FA 16 (3.7%) 5 (31.3%) 11 DXM 133 (31.1%) 52 (39.1%) 5 TMC IVI 21 (4.9%) 10 (47.6%) 1 TMC IVI + DXM 47 (11%) 27 (57.4%) 7 DXM + FA 37 (8.7%) 21 (56.8%) 6 TMC ST + DXM 33 (7.7%) 17 (51.5%) 4 TMC IVI + DXM + FA 6 (1.4%) 4 (66.7%) 8 TMC IVI + TMC ST + DXM 3 (0.7%) 1 (33.3%) 8 TMC ST + DXM + FA 1 (0.2%) 1 (100%) 8 TMC IVI + TMC ST 1 (0.2%) 0 (0.0%) 8 G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 429μm (± 129) and 386μm (± 144) in subjec s a e e inal ein occlusion and 497μm (± 108) and 356μm (± 125) a u ei ic pa ien s. In aocula p essu e (IOP) An IOP was measu ed a e e y isi , wi h an a e age o 14.3 (SD ± 3.3) mmHg be o e he i s s e oid adminis a ion, 15.7 (SD ± 4.6) mmHg a e 6–8weeks, 15.8 (SD ± 4.6) mmHg a e 6mon hs, 15.3 (SD ± 4.0) mmHg a e 12mon hs, 15.2 (SD ± 4.6) mmHg a e 18mon hs, and 15.5 (SD ± 5.4) mmHg a 24mon hs. S e oid esponse I a subjec p esen ed wi h an IOP ise o ≥ 25mmHg, we documen ed he ime a e i s s e oid adminis a ion and he he apy adminis e ed. Ou o a o al o 428 eyes, 168 p esen ed wi h an IOP ele a ion o ≥ 25mmHg, wi h a mean alue o 29.7 (SD ± 5.6) mmHg. The ise o ocula p essu e occu ed on a e age a e 7.5mon hs (SD ± 6.2) om he i s s e oid adminis a ion, wi h a median o 5.5mon hs. Subjec s p esen ed wi h an ele a ed ocula p essu e o ≥ 25mmHg a e a minimum o 2weeks and a maximum o 24mon hs. When i came o he ise o IOP a e adminis a ion o s e oids in a mono he apy o as a d ug combina ion, s e oids mainly leading o a ise in IOP included DXM (39.1% o all eyes ecei ing ha d ug), TMC ST (23.1%), TMC IVI combined wi h DXM (57.4%), DXM wi h FA (56.8%), and TMC ST wi h DXM (51.5%) (Table3 and Fig.1). Kaplan–Meie analysis and he Log Rank es showed a signi ican di e ence (p < 0.001) in be ween ypes o s e oidal medica ion used and he momen he pa ien had an IOP ele a ion. IOP ise was documen ed a e a median o 1mon h in case o TMC IVI and up o 11mon hs a e applica ion o FA (Fig.2). Addi ionally, 57.7% ou o all eyes wi h an IOP ele a ion we e ea ed wi h a s e oidal d ug mono he apy. Fu he mo e, he Fig. 1 Ele a ion o in aocula p essu e in eyes ea ed wi h mono he apy o combina ions o s e oidal medica ion (TMC ST = iamcinolone sub-Tenon; FA = luocinolone ace onide; DXM = dexame hasone; TMC IVI = iamcinolone in a i eal) Fig. 2 Median ime o an IOP ele a ion (in mon hs) a e he i s s e oidal agen applica ion in mono he apy, mos applied s e oid combina ions, and he es o s e oid combina ions (TMC ST = iamcinolone sub-Tenon; FA = luocinolone ace onide; DXM = dexame ha- sone; TMC IVI = iamcinolone in a i eal) G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 s e oid combina ions simila ly led o he IOP inc ease. Se en y-one eyes (42.3%) we e ea ed wi h s e oid com- bina ions ha esul ed in IOP ele a ion in 51 o 57% o eyes. The mos used combina ions included TMC ST wi h DXM, DXM wi h FA, and TMC IVI wi h DXM. Glaucoma pa ien s be o es e oid applica ion As a as eyes wi h p ima y open-angle glaucoma a e conce ned, we no ed 65 o hem. 30 (46.2%) had no IOP ele a ion a e applica ion o a s e oid agen and 35 (53.8%) p esen ed an inc eased IOP, which makes 20.8% o all 168 eyes wi h an ele a ed IOP in his s udy. Thus, eyes wi h p e ious glaucoma had a signi ican ly (p < 0.009) inc eased isk o de eloping an IOP ele a ion a e s e oid applica ion. Rubeosis i idis Rubeosis i idis, as an addi ional pa ame e ha may cause a ise in ocula p essu e, was documen ed in one pa ien be o e he s e oid he apy and addi ionally in wo pa ien s a e 6mon hs. The i s pa ien p esen ed wi h a diabe ic macula edema and p oli e a i e diabe ic e inopa hy; she also had an IOP ele a ion o 40mmHg maximum. The second subjec wi h he same diagnosis addi ionally had an ischemic maculopa hy, and IOP ose o mo e han 25mmHg. The hi d pa ien , who de eloped ubeosis i idis a e 6mon hs, had u ei is in e me- dia wi h asculi is, wi h IOP ele a ions o up o 50mmHg. Lens s a us – IOP ele a ion inphakic e suspseudophakic eyes Lens s a us showed o be s a is ically signi ican (p < 0.01) as a as ise o IOP is conce ned. Ou o 100 phakic eyes in gene al popula ion, 61% (n = 61) showed an ele a ed IOP a e s e oid applica ion. In compa ison, IOP ele a ion was documen ed only in 32.6% (n = 107) o pseudophakic eyes (n = 328) (Fig.3). Thus, lens s a- us could be a p edic i e ac o o IOP ele a ion a e s e oid adminis a ion. Gene al linea model (GLM) analysis o in luence o lens s a us on IOP ele a ion showed, ha i came slowe o IOP spike in phakic eyes, wi h a peak IOP o 18.4 (an a e age, ± 0.7) mmHg a 6mon hs. In pseudophakic eyes IOP spike o 18.0 (± 0.5) mmHg could be obse ed a 6–8weeks (p = 0.019, Fig.4). The apy In o al, 119 (70.8%) pa ien s ecei ed a conse a i e he apy wi h IOP-lowe ing eye d ops. The e was no s anda d scheme Fig. 3 Pe cen age o eyes wi h and wi hou IOP ele a ion in phakic and pseudophakic eyes (0 = no ise o IOP; 1 = ise o IOP) Fig. 4 A e age IOP in phakic and pseudophakic eyes (0 = no ca a ac su ge y; 1 = s a us pos ca a ac su ge y) G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 o p esc ibing he eye d ops, nei he in ou depa men no a ex e nal doc o ´s p ac ices. P esc ip ions included alpha agonis s, be a blocke s and ca bonic anhyd ase inhibi o s. When he ollow-up exams showed an IOP o ≤ 20mmHg, we conside ed his as well egula ed, and a conse a i e he apy could be discon inued. Su icien IOP egula ion was achie ed in 82 eyes (68.9%) ea ed wi h opical he apy. In 37 eyes (31.1%) wi h pe sis- en ly ele a ed in aocula p essu e, opical he apy had o be con inued pas he ollow-up. Su gical in e en ion was pe o med in 21 eyes wi h pe sis en ly ele a ed IOP and inc ease in cup- o-disc a ion (CDR). Ou o hese 21 cases, we pe o med a cyclopho o- coagula ion in 14 eyes (8.3%), a il e ing su ge y in 3 eyes (1.8%) and we emo ed a s e oidal d ug implan – FA – om 4 eyes (2.4%). S e oidal d ug implan s we e emo ed in 1 eye ea ed o a pos ope a i e macula edema (25%), in 2 eyes ea ed o a diabe ic macula edema (50%) and in 1 eye su e ing om a e inal ein occlusion (25%). 28 eyes ecei ed no he apy (16.7%). Discussion Common side e ec s o in aocula glucoco icoids may include pos injec ion in ec ious endoph halmi is, second- a y ocula hype ension, o seconda y s e oid-induced open-angle glaucoma, hegma ogenous e inal de ach- men , pos injec ion s e oid-induced ca a ac , cen al se ous cho io e inopa hy, and oxic e ec s [27–30]. Howe e , an inc ease in IOP is one o he mos widely discussed opics in e ms o i s possible u u e implica ions, i.e. de elopmen o a seconda y s e oid-induced glaucoma. When compa ed o in a i eal injec ion, sub-Tenon applica ion o TMC has a lowe isk o IOP ele a ion due o a lowe in aocula con- cen a ion and sho e du a ion o ac ion [31, 32]. Li e a u e desc ibes a hype ensi e ocula esponse in up o 50% o eyes ha a e injec ed wi h a co icos e oid [10, 33]. In ou s udy, 57.7% ou o all eyes wi h an IOP ele a ion had been ea ed wi h a s e oidal d ug mono he apy. Acco d- ing o he o icial summa y o p oduc cha ac e is ics, an IOP ele a ion o ≥ 10mmHg om baseline occu ed in 28% a any isi , while an IOL o ≥ 30mmHg occu ed in 15% o pa ien s who we e subjec s o an obse a ional s udy and ecei ed DXM. In ou s udy, an IOP ise, de ined as a single measu emen o ≥ 25mmHg [12], was documen ed in 39.3% o all eyes. Simila ly, Rezkallah e al. [34] analyzed 494 eyes a e an in a i eal injec ion o a dexame hasone-implan , ou o which 32.6% p esen ed wi h ocula hype ension (OHT). The main indica ions o ea men in hei s udy we e e inal ein occlusion, diabe ic macula edema, pos - su gical macula edema and u ei is. The SAFODEX s udy [12] showed ha dexame hasone-implan s, injec ed in 421 eyes o e inal ein occlusion, diabe ic macula edema, pos - su gical macula edema and u ei is, caused OHT in 28.5% o cases. In he ZERO s udy, conduc ed by Schmi z e al., less han 20% o he eyes a e dexame hasone injec ions o e inal ein occlusion p esen ed wi h ele a ed IOP. As a as in a i eal injec ions wi h luocinolone ace o- nide a e conce ned, main ad e se e ec s include he isk o aised in aocula p essu e, as well as ca a ac de elopmen [23]. Acco ding o ILUVIEN’s summa y o p oduc cha - ac e is ics, IOL ele a ion is a e y common ad e se e en wi h IOP > 25mmHg a 21% and > 30mmHg a 14% o pa ien s in an obse a ional s udy. 38% o subjec s equi ed an IOP-lowe ing medica ion, whe eas 5.6% had su ge y o lowe IOP. In ou s udy, 5 o 16 eyes (31.3%) wi h FA mono- he apy had an ele a ed IOP which occu ed a e a median o 11mon hs. In he FAME s udy, IOP inc eased in 37.1% o eyes a e ecei ing an FA-injec ion [23]. In compa ison, Al aqawi e al. [35] documen ed an IOP ise o ≥ 10mmHg in 3 eyes (11%). Fallico e al. [36] epo ed ha 27% o hei subjec s equi ed IOP-lowe ing d ops and 3% glaucoma su - ge y. In e es ingly, 1.8% o ou pa ien s’ eyes unde wen a il e ing glaucoma su ge y and 2.4% had a s e oidal d ug implan – luocinolone ace onide in all cases – emo ed. In an a icle by Tao e al. [28], abou 40% o eyes de el- oped a seconda y OHT a e an in a i eal iamcinolone he apy, whe eas in ou s udy 47.6% o eyes p esen ed wi h ocula hype ension. Fu he mo e, sub-Tenon iamcinolone ace onide caused an IOP ele a ion in 23.1% o eyes. Handzel e al. [37] as well as Ca dillo e al. [31] s a ed, ha he e was no signi ican ise in IOP a e sub-Tenon iamcinolone ace- onide, wi h Ca dillo p esen ing no di e ence be ween he 2 ypes o iamcinolone ace onide applica ion (in a i eal e sus sub-Tenon). The mos common s e oid combina ions ha esul ed in an inc ease in IOP we e in a i eal iamcinolone ace on- ide combined wi h dexame hasone (57.4%), dexame hasone wi h luocinolone ace onide (56.8%), and sub-Tenon iam- cinolone ace onide wi h dexame hasone (51.5%) (Table3 and Fig.1). These esul s le us suspec ha he apy wi h dexame hasone, ei he as a mono he apy o in combina ion wi h ano he s e oid, ends o inc ease IOP. The ise o ocula p essu e occu ed on a e age a e 7.5mon hs om he i s s e oid adminis a ion, wi h a median o 5.5mon hs. Subjec s p esen ed wi h an ele a ed ocula p essu e o ≥ 25mmHg a e a minimum o 2weeks and a maximum o 24mon hs. IOP ise was documen ed a e a median o 1mon h in case o in a i eal iamci- nolone ace onide, wi h he i s IOP ele a ion occu ing a e no mo e han 2weeks. This co ela es wi h esul s by Jonas e al. [30] who showed IOP ise al eady du ing he i s week a e he in a i eal injec ion. The longes mean inc ease o IOP wi h 11mon hs was obse ed a e he applica ion o luocinolone ace onide. In compa ison, Al aqawi e al. [35] G ae e's A chi e o Clinical and Expe imen al Oph halmology 1 3 documen ed an IOP ise a e a mean o 3mon hs a e he applica ion. In ou s udy, in 28 eyes (16.7%) wi h a single IOP ise, he e was spon aneous no maliza ion o IOP a e discon- inua ion o s e oidal he apy, wi hou p essu e-lowe ing opical he apy. Gi en a s e oid induced ele a ed IOP, no - maliza ion is possible a e simply s opping he s e oidal he apy. Howe e , o p e en s uc u al and/o unc ional de e io a ion due o con inuously inc eased IOP, pa ien s su e ing om SIOH, o SIG can be success ully managed wi h opical glaucoma medica ion, jus he same as pa ien s wi h p ima y open-angle glaucoma and ocula hype en- sion, wi hou he impac o s e oids. I hese ea men s ail o con ol he IOP, pa ien s mus unde go a adi ional glaucoma su gical in e en ion, which is only equi ed in less han 2% o cases [30]. Acco ding o he o icial sum- ma y o p oduc cha ac e is ics o OZURDEX®, opical IOP lowe ing medica ion was applied in 42%, while su gi- cal in e en ion o ele a ed IOP was needed in 1.2% o pa ien s who we e subjec s o an obse a ional s udy [38]. In he MEAD S udy o e 40% o eyes equi ed a opical an iglaucoma he apy and 0.3% o eyes we e ea ed wi h incisional glaucoma su ge y [39]. Mo eo e , he Re ise - FA implan s udy epo ed an IOP ele a ion o ≥ 30mmHg in mo e han 60% o eyes. As a esul , one hi d o hese eyes equi ed glaucoma il a ion su ge y o explan a ion o he de ice [40]. Ou s udy showed 168 eyes (39.3%) wi h an OHT a e s e oid injec ion, which we e ea ed conse a i ely wi h opical an iglaucoma he apy (70.8% o eyes wi h ele a ed IOP). In compa ison, 10% o eyes equi ed opical glaucoma ea men a e an IOP ele a ion in he EMR-s udy [41]. We conside ed an IOP o be well con olled when eaching an IOP o ≤ 20mmHg unde op- ical he apy, which was achie ed in 82 eyes (48.8%) a e 4.5mon hs on a e age. Li e al. [42] eached IOP con ol simila ly by mon h 4. Su gical in e en ion was needed in 21 eyes (12.5%) wi h cyclopho ocoagula ion pe o med in 14 eyes (8.3%), glaucoma il a ion su ge y pe o med in 3 eyes (1.8%) and s e oidal d ug implan emo al in 4 cases (2.4%). The IRSS S udy showed ha 2% o eyes equi ed incisional glaucoma su ge y [43], whe eas in he EMR- s udy, his was only necessa y in 0.3% [41]. As a as eyes wi h p ima y open-angle glaucoma a e conce ned, 35 ou o 65 (53.8%) p esen ed wi h an inc eased IOP, which is 20.8% ou o all 168 eyes wi h an OHT and showed o be s a is ically signi ican wi h a p- alue < 0.009. Simila ly, in a s udy by Chin e al. [44], ou o 13 eyes wi h p e-exis ing p ima y open-angle glau- coma, 6 de eloped OHT (46.2%). Mo eo e , Le in e al. [45] analyzed pa ien s wi h a his o y o co icos e oid induced IOP ele a ion and his o ical non- esponde s a e sub-Tenon co icos e oid injec ions. Thei esul s showed ha a highe a e o ecu en IOP ele a ion de eloped in his o ical esponde eyes (44%) compa ed wi h 13% in non- esponde s. Lens s a us migh be a p edic i e ac o o IOP ele a ion a e s e oid applica ion, as IOP inc eased in 61% o phakic eyes, compa ed o 32.6% o pseudophakic eyes. One o he majo limi a ions o ou s udy is he e o- spec i e design which is he cause o some missing da a. The e o e, p ospec i e designs should be c ea ed o be e esea ch on and unde s and his opic. Mo eo e , as ou clinic is loca ed in a u al a ea wi h a poo ly de eloped pub- lic anspo sys em and some o he pa ien s a e mul imo - bid, he connec ion o a egula check-up, which is decisi e o he success o il e ing ope a ions, was no gi en. Fo his eason, we p e e ed cyclopho ocoagula ion because o he educed need o pos ope a i e ollow-up. Conclusions In conclusion, e e y s e oid he apy, ega dless o he agen o indica ion, can lead o he de elopmen o ocula hype - ension. The isk may a y acco ding o he s e oids and he ea men (e.g. mono he apy e sus d ug combina ion) used. Resul s o ou s udy le us suspec ha especially he apy wi h in a i eal dexame hasone, ei he as a mono he apy o in combina ion wi h ano he s e oid, ends o inc ease IOP mo e han o he s e oids. An ele a ed IOP p edisposes he pa ien o de elop seconda y s e oid-induced hype ension, as well as seconda y s e oid-induced glaucoma wi h all o he disease’s sigh - h ea ening aspec s. Each pa ien should be in ensi ely in o med abou he possible isks o he apy and a end egula IOP con ols du ing ollow-up o ini ia e an ea ly in e en ion i necessa y. Funding Open Access unding enabled and o ganized by P ojek DEAL. No unding was ecei ed o his esea ch. Decla a ions E hical app o al All p ocedu es pe o med in his s udy we e in acco dance wi h he e hical s anda ds o he Saa land Uni e si y Medi- cal Cen e and he E hics Commi ee o he Medical Associa ion o Saa land, Ge many (N . 123/20, da e: 16.06.2020), as well as wi h he 1964 Helsinki decla a ion and i s la e amendmen s o compa able e hical s anda ds. This a icle does no con ain any s udies wi h animals pe o med by any o he au ho s. In o med consen Fo his ype o s udy o mal consen is no equi ed. Con lic o in e es All au ho s ce i y ha hey ha e no a ilia ions wi h o in ol emen in any o ganiza ion o en i y wi h any inancial in- e es (such as hono a ia; educa ional g an s; pa icipa ion in speake s' bu eaus; membe ship, employmen , consul ancies, s ock owne ship, o o he equi y in e es ; and expe es imony o pa en -licensing a -