Posts Tagged ‘primary pci’

Interventional cardiologist extraordinary  cath lab tips invasive great

Read Full Post »

Multivessel PCI during acute STEMI is forbidden except in cardiogenic  shock . (or in some very unstable patients without cardiogenic shock)

The reason

  • During acute MI   hemodynamics  are precariously balanced.We do not know yet how  emergency multivessel plasty alters this .
  • Our  initial aim should be   confined to myocardial salvage in the IRA . Total myocardial revascularization is niether  the  priority nor its desirable.
  • The more  time  you spend  within the inflamed coronary artery , more its  hazardous.
  • Multiple stenting  is prone for thrombus   and  migration  into side branch .
  • Stent opposition is sub optimal in many thrombus infested lesions.

Still  . . .  in real world it is extremely difficult to curtail the urge to stent  all eligible lesion during primary PCI !

multivessel angioplasy during stemi

How to avoid it ? 

If the patient is poor or the insurance limit is low , the issue  of multi vessel stenting does not arise at all  !

Always  ignore  complex  non IRA lesions  during primary  PCI. Be happy if a non IRA has a bifurcation lesion !

Still , some lovely looking lesions in non IRA  would be  tempting  and inviting .  Indulge at your own risk !

* Please remember if  the proximal  LAD  has a non IRA lesion , it may be sensible to attempt  simultaneous revascularisation even if the patient is stable !

Other unrealistic advice

  • Keep the professional fee and other benefits   fixed whether  we do a single or multiple   vessel stenting (Realise  . . .  surgeons do not charge more for a  4  vessel by-pass graft  than a single  ! )
  • Keep the current AHA/ACC/ESC guidelines pasted right next to the fluroscopy monitor .
  • Ask your subordinates to repeatedly caution   you  about the possible  excesses and ask them to wave a red flag !
  • You may  empower the   senior staff nurse   with a veto power  to shut off the cath lab once IRA plasty is  completed and the patient  is stable.
  • In extreme  situations , keep a cath  marshal ready to manually evacuate  the primary operator  from cath lab !


multivessel angioplasty during stemi


Read Full Post »

For STEMI management there are  6 management protocols available

  1. Thrombolysis
  2. Primary PCI
  3. Rescue PCI
  4. Facilitated PCI
  5. Pharmaco -Invasive approach
  6. CABG

*CABG is rarely used except in  severe mechanical complication.

There is some  issues in differentiating  facilitated PCI and  Pharmaco Invasive Approach.

What do we facilitate ? How we do it ?

PCI in acute STEMI is done in a thrombotic milleu. So we get sub optimal results .Hence to facilitate it we try using

either 2B-3A antagonists, Newer Heparins, or even thrombolytic agents before submitting them for PCI

Where is this facilitation done ?

Facilitated PCI is done in small hospitals where  there  is no cath lab or cath lab is available only during office hours.

Facilitation can be done in either in same hospital or on the way to big hospital

Is there a time window to start  this ?

The main aim was to was to facilitate the PCI .Hence time window was not considered vital in few studies (Wrongly though !) ideally it should be started as early as the first contact . Since facilitation can be started earlier the time window is 0-24 hours .

What happened to the concept of f-PCI ?

It died a premature death  and  last rites were  completed when the FINNESE trial was out .

But it left behind a daughter concept ie in selected patients if the facilitation is done early , especially in those patients who are going to get the subsequent PCI late ,or in high risk individuals  , the initial  pharmacological facilitation* was indeed useful.)

*If  facilitation was with   fibrinolytic agents (Not 2a/2b )  .It is very important the benefits of facilitation is mainly  attributed to the time gain in achieving partial opening of IRA  making it more complete salvage of the subsequent PCI .

This aspect later on named as PIA .

Pharmaco- invasive approach(PIA)

We know p PCI is a race against time .We also  know fibrinolytic therapy  fares well in this race  but   pPCI  beats in   effectiveness  .

So what prevents us to combine the swiftness the fibrinolysis and the robustness of pPCI ?  That is  like getting the best of both world .( It is not that easy thing accomplish after all 1+1 in medicine is rarely 2 !)

In it’s core principle it  is same as f-PCI . But facilitation is done only with fibrinolytic agent (Not 2B-3A) . Pharmaco Invasive strategy can be started in any small hospital/ In the ambulance /. It  is routinely followed by PCI whether the initial thrombolysis is successful or not . PIA should not be done before 3 hours window if  a timely pPCI is feasible.  Hence PIA has a typical time window of 3-24 hours .


f-PCI is combining  various anti-platelet and fibrinlytic strategy prior to PCI . It was found  to be useless if it is used routinely in all cases of pPCI. (Rather 2B-3A  was useful  if  only the facilitation was done within the cath lab to prevent procedure related issues) .Time window can be between 0-24h .

Pharmaco Invasive approach (PIA)   is actually a type of f-PCI where  fibrinolytic agents are used routinely which is followed by mandatory angiogram and PCI in all deserving cases.Many still  believe the facilitation in PIA is primarily accured in  shortening the   time to reperfusion  rather than altering the thrombus load and morphology  ! Time window is usually between 3-24 hours.

Read Full Post »

Primary PCI is presumed to be the ultimate  , undisputed reperfusion  strategy  in STEMI .  Still , time and again one study or  other strips down  this   “Numero Uno”  status of pPCI  .  If it is really supreme ,  such awkward  situation shouldn’t arise  too often . More importantly , the  major reason for  dubious real world record of  pPCI  goes beyond  the time and logistic factors (which is considered the only issue  for pPCI by most interventionist ! ) There is something more to it that is invisible ! (Is it the no reflow ?)

The nearly flawless study from Belgium ( STREAM Just released in ACC 2013/Sanfransisco ) , pre-hopsital or early fibrinolysis has proven to be superior in the prevention major end points at 30 days .

  1. Death
  2. Re-infarction
  3. CHF


The major surprise was pre-hospital  fibrinolysis  showed less  incidence of cardiogenic shock . ( pPCI

group had more of this ( 4.4 VS 5.9 %  in STREAM )

Now . . .  shall I make a provocative statement ?

while pPCI may be treatment of choice for cardiogenic shock . . . but it may  also confer a risk of cardiogenic shock in otherwise low risk MI !

Caution  and  conclusion

STREAM population applies strictly to 1 to 3 hour time window . It does not apply to either before or after that ! Simply put,we do not have  guts to compare fibrinolysis and pPCI  in patients who arrive  within one hour into a facility where 24 hour cath lab facility is available .  We call it unethical to do a study like that !  I personally feel it is really unethical  if we do not do a study in this time frame . The reasoning is  simple and very personal .In a  large  Government  hospital   where  we do not have primary PCI program  our net mortality for STEMI never exceeded 7-8 %  over a period of 10 years  , Which  is almost at par with global data on pPCI. (Our door to needle time is an unbelivebale  8-12 minutes ! that  too only streptokinase !)

Adding Further controversy

pPCI  is indeed a superior reperfusion strategy . No one can dispute that .But its superiority  is not  realised  in  every patient  who gets it.  The benefits are accrued if and only if it is  used most judiciously . In Low risk , small regional  , branch vessel STEMI ,  pPCI has never been  shown superior . It is well recognised ,  upto 15 % of STEMI is likely to spontaneously abort or experience very good spontaneous recannalisation . By rushing these  patients very early into cath lab pPCI   meddles with the natural anti fibrinolytic mechanisms . It is this population who  invite all the procedural hazards. .

Is this the reason STREAM had  more  cardiogenic shocks in pPCI limb ?

I think STREAM has  strengthened the case in favor of fibrinolysis in this  ever ending debate .

I would  seriously believe  pPCI is hanging it’s superiority over fibrinolysis with a wafer thin mortality advantage . pPCI may  not be recommended in a routine fashion to all STEMI  population even if they arrive within 6 hours and able to perform the plasty fast .  Science is   . . .  after all . . .  continuing  confrontations with our  assumptions !

Counter point

STREAM is not an exclusive study comparing fibrinolysis and PCI . It is a  study comparing   Pharmaco Invasive approach vs  pure invasive approach . 80 %  of patients in the  fibrinolytic limb ultimately received PCI and  stenting . It simply doesnot make sense to conclude fibrinolysis is superior to PCI . Most of the beneficial  effects on 30 day outcome may reflect the timely PCI  in the lytic group.


Read Full Post »

A patient with  extensive anterior STEMI  presented 18 hours  after onset of  chest pain . He  was  other wise stable and free from angina but had persistent ST elevation (5mm in V 1 to V 5 ). He had a  total occlusion of LAD  with TIMI zero  flow . He had a  tight PDA lesion  as well . A bed side echo revealed LV EF  of 50% . The septum was hypo-kinetic but did not appear severely dysfunctional .

So , it was decided to open up the LAD. The moment  LAD was opened he developed severe acute LVF  /   flash pulmonary edema   .  Even after a 30 minutes of  heart (Fire )  fighting  he could not be resuscitated .

What is the mechanism of death here ? Expert  STEMI interventionist  from core  labs  may answer this !

An acute ischemic MR with myocardial disruption was suggested . Why it  was triggered after opening the IRA ?

Three mechanisms were discussed

  1. Re-perfusion injury
  2. Collateral  damage
  3. Physiological  de-stabilisation of  Contra -Lateral lesion (Remote lesions )

Re-perfusion Injury ?  How relevant it is in cath lab ?

Is re-perfusion injury  electrical  ,  mechanical or  both ?

In this particular patient even though there was a total LAD occlusion , the segments supplied   by  the LAD  was partially functional and  it was contributing to LV  pump function.  The moment  a trickle of   flow was established  , some thing happened and the whatever  little mechanical function  his LV  had  was also interrupted  . The LV came to standstill and the patient died .

If re-perfusion Injury is  simply an   electrical  event   like VF ,  it can be resuscitated . If it is mechanical  outcome is bad ! This is not a new concept  . It is  part of the  once famous  concept called myocardial stunning . There are  lots of reasons   for stunning  to be a  clinically relevant phenomenon .Unfortunately   if any cardiologist talks about it in 2012 ,  he is at risk of  labeled  as old fashioned !

Collateral damage.

One more mechanism which we feel that  might have contributed to death here  is   the  “collateral damage” .(This is not cross fire !)

We know collaterals can be recruited within 12 hours in many STEMI patients . In some  it can even salvage  significant mass of  myocardium . The acute collaterals to LAD may be interrupted  during primary PCI . Once you poke the lesion the coronary  vascular  bed which had dilated  (as a response to total occlusion ) may react with inappropriate vasoconstriction . This raises the local hydrostatic pressure (Myocardial edema)  and further impede  the   incoming  micro collateral flow . This a very  critical time  for the myocardium  where antegrade and retrograde flow are kept in a fine balance .

Interference with remote lesion Hemodynamics .

Another possibility  is  the  opening the  LAD lesion some how  impact on remote lesional  flow as well (PDA  in this patient  )

Please remember ,

Even a transient hypo- tension can have  devastating effect in  the  hemo -dynamics   of  non IRA  territory  especially if it harbors a critical lesion !

Final message

Coming to the title question  , Is no – flow better than  slow- flow in late presenters of STEMI ?

Common sense dictates whenever  an artery is obstructed  just get rid of it.  When  it  comes to the heart it must be done in an urgent basis That is the essence of primary angioplasty  . . .  agreed . But in this  patient  I believe ,  the  common sense  was proved wrong !

Truths are always hidden.  The  science of  myocardial re-perfusion is a perfect example . We need to learn a lot still !

This I  call as  Para cardiology : Heart  facts without  evidence !

Counter point

One may argue this   is an  exceptional case  in STEMI  intervention. Don’t  hype   exceptions  and undermine the importance of a great concept ! Exceptions  and rules  are directly related to our  experience  we have accrued.  Exceptions are the great  knowledge substrates  and help  crack  medical  mysteries !

Read Full Post »

The key word for  successful  primary PCI  is

  •  Suction &  Aspiration of thrombus  with   micro catheters like  export catheters
  • One can do away with a stent during primary PCI but can never do away aspiration
  • Distal protection as concept is rapidly dying out as we aim to remove all the thrombus .

Tips for effective thrombus aspiration

  • Apply continuous negative pressure once catheter reaches the thrombus do not release  it till you enter back into the guide.
  • Make sure  you are sucking only  blood  products  not the  endothelium
  • Watch out for  side branch spill over.
  • 7F sheath 7F catheter ideal for aspirating  with a  micro catheter
  • Please be informed some thrombus require more negative pressure especially  in the late  presenters of STEMI

* During dire emergency when you do not have a specialized suction catheter do not hesitate to push  even a diagnostic catheter into the coronary .We have  saved few lives !

Crazy   questions  in primary PCI  ( or Is  it futuristic )

Can we connect the suction apparatus into LAD micro catheter ?

Do we have camera guided suction catheter ?

Can you flush the thrombus if you are not succeeding in aspiration ?

Is ultrasonic desiccation  of thrombus possible ?


Some of the tips were  gathered from the recently concluded  India Live  2012  conference   in New Delhi .

Read Full Post »

A middle-aged man was rushed to cath lab with  extensive antero- lateral STEMI   . Primary angioplasty  was planned.The coronary angiogram  showed a critical LAD and a  total LCX lesion just beyond the bifurcation . Both lesions looked irregular and hazy . RCA had insignificant lesions . Patient was  stable hemo-dynamicaly .

The moment  we  saw the angiogram,   we knew ,  we had a real problem on hand !  . First of all ,  It looked a complex lesion for  a pPCI .(A brief  thought about an emergency CABG  creeped in,  but was dropped  with   enthusiastic residents voted unequivocally in favor of  PCI . Of course , to be frank  we didn’t have a  CABG team ready either ! )

So the plan  was : To open the IRA  . . .  &   forget the non IRA (  for the time being )  which  is the current management mantra as on 2012 !

Trouble from unusual  quarters !  By the way   . . . which is the IRA  ?

Even as the consultant  was  initiating  the rituals with wires and balloons  to tackle the LCX , some one behind the consultant  mumbled  “why can’t the LAD be  the IRA ”  After all  , it  also has  a critical lesion  and  mind you  we are dealing a case of   anterior MI  !) . That mumble  was  loud enough to create buzz of confusion in cath lab .

Now everyone quipped  , ” IRA is  what ?

Is that  the critical mid LAD lesion ?

(or ) Is it the total LCX   (or ) Both ?

Logic would suggest in the setting of STEMI   any  total occlusion should be considered as IRA  . Of course  , one can not be dogmatic about it.  When a patient is having   anterolateal MI  both LAD  and  LCX can contribute to the MI .

What about  proposing a new  concept  of  “Double IRA” ?

When  multiple  plaques  are  activated suddenly in unstable angina ,  it is  possible  for multiple IRAs  to occur  in STEMI as well . But this issue is rarely discussed in literature .

Other  possibilities

The 100 %  lesion  in LCX  could  still be the primary culprit  and  a  thrombus migration into LAD   might have  resulted  in  infarct  extension into anterior wall .

Further ,  confounding may occur if a patient with chronic total occlusion  develop a  SEMI  . It  makes it really difficult to identify the  IRA.

When  the supposedly gold standard coronary angiogram   fails to identify the IRA ,  what shall  we do ?

Go to the basics . The good old ECG might help .

(Please beware in a patient  with pre- existing  multi-vesel CAD  , none of  the ECG algorithms work  to localise  IRA !(Especially   the famous Wellen’s miserably fails ! )

Still unclear ?  Look for the wall  motion defects  in echo . An echo cardiogram (Need to be meticulous )   will help match the  dysfunctional segment with IRA.

Wall  motion defects are notoriously  error prone in ACS  for  two reasons.

  • We do not have easy and accurate  methods to differentiate ischemic wall motion defect from infarct related wall motion defects.
  • Tethering artifacts  ,  differential behavior  epicardial  vs endocardial ischema on contractility   will confound  the issue .

So what is left ?

One  need to  go back  to the CAG again  . Have  a critical  look  at the lesion once more. Look for thrombus or eccentric /unstable lesions . If  present it is  going  to be the IRA in 90 % of times. Let it be a  wild guess in the remaining 10 % .

There is also a  practical solution . Poke the lesions  with your favorite  guidewire ! . The one that  gives way easily  is likely to be the  IRA !

Finally,  if the confusion still prevails ,

Stent both the  lesions. That’s what , was  done to this patient . Many would have thought ,  this should  have been the default approach instead of scratching  our heads to identify the IRA ,  wasting crucial minutes !

Final message  : 

Current  guidelines do not recommend  pPCI for non -IRA   at the same sitting  of  IRA pPCI . However the issue  of  IRA  “too close to call is  rarely addressed.I do not know  how commonly  this issue is encountered  in angiographic  core labs that deal  huge loads  of pPCI world wide .

Our  early  experience  suggest  the problem is   real ,  unique and  definitely not rare  .

What is your take ?  We argue guidelines committee   to  specifically  address  the issue of  uncertain IRA as a  branch point in the pPCI decision making  tree !

Read Full Post »

A patient with  chest pain   is being rushed   into the ER  of  a medium sized cardiac facility in an urban county of India .The fellow briefs the senior consultant cardiologist about the arrival of  this  extensive anterior  STEMI  over phone.  The  consultant  enquires  about the “symptom to door time” and  was little concerned  when the fellow said ” it is only  2 hours sir”  .The fellow was amused with the consultant’s  reaction.

The consultant arrived in 15 minutes  and  began  the all important  discussion with the  patient  and his spouse  .(Meanwhile tablet clopidogrel and Aspirin was  loaded  per orally . Note : Heparin is not given yet )

Cardiologist : It is a  massive heart attack . One of your coronary artery  got  blocked suddenly , I have to remove the block at the earliest . There are  two ways of doing it .  One is fibrinolysis which lyses the clot .It can open up  your vessel  , though incompletely but would prevent myocardial damage  at the earliest .

The other one is PCI ,  which if performed rapidly  will completely  open up the vessel in question , what  we call TIMI 3 flow ,  But it has to be done within one hour.

Patient : Which is better doctor for me ?

Cardiologist : My cath lab  logistics  does not allow me to do a  PCI within the stipulated  time ,   still  I wish  to perform   a primary angioplasty  as i am not a believer  in thrombolysis !

So , even though you are  eligible for both modes of re- perfusion, in the strict sense doing a  PCI  which will  ultimately be  delayed  beyond the recommended time window  is technically contraindicated “

(Please  be notified : Currently , if the delay is during the procedure due to some technical issues after starting the procedure  or if the delay is at  the patient   level in arranging the required finance or insurance clearance it can be condoned without any ethical issues )

Patients spouse : Doctor please save my  husband . You  start  the fibrinolysis  every minute is risky isn’t doctor , and do the PCI once its ready doctor .

No . . . Fibrinolysis  does not combine well with PCI  in fact it will worsen the situation .   Thats what FINNESSE  study says . We can do only one of them . . . not both .

The patient and spouse (Terribly confused  by now )

Cardiologist:  By the way , what is  your insurance limit  sir ?

Patients spouse : Its 4 lakhs

Cardiologist : OK , that should be suffice  90 out of 100 times . Any way keep another lakh ready in case I need IABP.

Patient : My pain is worsening doctor at least relieve  my pain till this debate  is over !

(A patient’s relative  browses his  i phone  and argues  the  cardiologist   to administer streptokinase at the earliest . Suddenly  the patient family lobbies for fibrinolytic mode  empowered by the i phone guy )

One of them shouts “Please doctor you do something either take him inside cath lab or  start a fibrinolyitc therapy “

How did the cardiologist  fight his way through ? ( He gulps a cup of coffee and  starts a fresh discussion)

Cardiologist : 
I am sorry , you have come too early  for PCI .  Guidelines do not allow me to choose PCI in the first hour  as our  anticipated delay for PCI is more than 3 hours . I wish you arrived after  the 3 hour window .

“Of course you are on time for thrombolysis”  which  unfortunately I am against !

Now, I am going to wait ( You may think it is a waste of time i call it as patient preparation time )  .This waiting period incidentally  allows us to cross  three  hour  time  window ,  then the issue of not lysing  does n’t creep in at all . A PCI done after 3 hours is not a race against time , while a PCI done early is like  a power play in cricket .

I do not know whether this delay which is happening  right now –  Intentional /Unintentional,  Scientific / unscientific reasons  would  damage your heart or not !

Since you have an extensive MI , I earnestly believe  with all my wisdom and knowledge, you will do well with primary PCI  even if  i do it  little late .  Please allow me to violate the standard criteria in the interest of your heart .

Cardiologist : I wonder , if you had arrived little late we wouldn’t be discussing this terrible conversation at all .Your myocardium  will also feel thrilled for getting a better mode of re-perfusion. You put me in an awkward situation by coming early !

Patient : But  . . . doctor every one  tells me ,  one should reach the hospital  after a heart attack  at the earliest  is it not ?   Please believe me doctor ,  I  made  extraordinary  efforts to  arrive early to your hospital , but you have put me on hold doctor !

Cardiologist :I agree  . . . but you won’t understand the modern jargons  in  interventional  cardiology ,  some times (Or is  it many times !) we will be doing the diagonally opposite to what  is   preached  in text books  .  Both intentional and unintentional delays are common  in the emergency cardiac  care . Do not bother about it   . . . science will take care of it !

Patient : You mean  . . . you are going to waste the golden hour in STEMI by simply arranging for cath lab staffs and machinery .

Cardiologist : You got the point right ! Just sit back enjoy the  flight !

Patient : Your wish is my wish doctor . Please handle with care doc !

* Please note this is an imaginary conversation ( Most often happens in silent mode !) in many of the cath labs which do not have 24 hour service. In a country  with  100 crore population like India ,  less than a dozen cathlabs  work round the clock . Guess  how often such a  situation would come in day to  day cardiology practice.

Read Full Post »

During   primary PCI , the weakest link  for a  cardiologist  is  , he is never sure whether  the  metal jacket  has covered the entire  disease segment with optimal apposition .  (Geographical miss is another issue !)

This is because  , even though the inflation pressure is  uniform  within the balloon ,  the required  apposition pressure is not the same .This is obvious as the lesion surface has a varying consistency and uneven surfaces . It is a  huge guess to quantitate the relative  contribution of thrombus and plaque  within  the 100 % occlusion  that has resulted in the STEMI. Hence  some areas may get over apposed and others lesser apposed. Further , the stent -vessel wall interface  in all likely hood enclose a   layer  of clot .This is almost certain  during complex primary PCI. One can imagine the sequel if this thrombus layer dissolves later ! (Edentulous stent )

It is surprising , why cardiologists has  so far not  thought  of a  self expanding stent  which  can snugly appose the vessel wall in this setting  . The   radial strength   from the  stored potential energy can be used up future use. This is most important  in first few days following STEMI  , when the coronary arterial lumen can vary depending  upon the

  • Vasomotor  tone .
  • presence of thrombus
  • Plaque   ploughing /milking  effect
  • Vascular remodeling

Cardiologists  deploy a stent  based on the morphology  on day zero of STEMI  .This may be  totally irrelevant  , since after a  few days    the lesion may change its morphology ,  thrombus may migrate , vascular  dimension may change. In such a  situation*  , a self expanding stent can tackle these issues very effectively by constantly adjusting  and fine tuning the luminal  diameter and  the apposition pressure . It  does not give any chance  for  thrombus to form  between the vessel wall and stent .

Here is a study that gives fresh insights regarding the role of self expanding stents during STEMI .

Note the “Auto adjusting”  of stent diameter  in the first few days after  the stent deployment, depending upon the luminal needs !



* Logically  during  primary PCI for  STEMI  ,  POBA and thrombus suction  may be the best option in many as all stent related complication is instantly eliminated .But it is a battered concept ,  most of the current day cardiologists would feel guilty to come out of  the cath lab  without a stent  in  primary PCI scenario  !

Final message

Self  expanding stents during primary PCI :  Is it a  perfect solution  for optimal stent apposition  ?

It seems so  . . . but  the track record of current cardiology devices never fulfilled the initial promises !

Read Full Post »

Do you fear to deploy DES during primary PCI ?

  1. No. Not at all !
  2. May be  “Yes”
  3. Yes, definitely
  4. It depends upon which drug it Elutes !

Answers  that might  decode  the  cardiologist  mind. ( Excuse me  if it errs !)

If the answer is 1 ,  You are the most optimistic cardiologist  and scientifically  updated ,   data driven  cardiologist , and with little concern for the patient welfare.

If the answer is 2 , Your are a cautious cardiologist may be . . . may be . . .  an   ideal one . The chances of you , using a  DES is  still low in  STEMI.

If the answer is 3 , You are a pessimist and  with   anti technology thoughts but still you  have more concern  for  your patients!

If your answer is 4 ,  You are undecided  and probably  ignorant as well . Most likely   you would not use it  for some reason  .You need to read more  on the topic .

By the way , my answer is  response three.

Read further ,  the EXAMINATION trial just released in ESC 2011 Paris .

Read Full Post »

Older Posts »


Get every new post delivered to your Inbox.

Join 349 other followers