THE CURSE OF THE CERVICAL COLLAR

Cervical CollarFor many years, ATLS has dictated cervical collar as part of the A in ABC, and any patient that enters a trauma bay gets a cervical collar slapped on before anyone cares about airways, breathing and circulation. The last couple of years, some rougue docs have tried opposing the validity of the extreme focus on cervical collars, and it is finally starting to trickle into the system. Here’s the case against cervical collars – and for bringing the focus back on the important parts of the ABC to save your patient.

The cervical collar dogma
The cervical collar has become a curse. It’s seen as the shining proof of good quality trauma care. No ambulance service dare deliver a patient to a trauma bay without a cervical collar. Not based on trauma mechanism or patient symptoms, but solely based on fear of criticism from the in-hospital ATLS-trained trauma team leader. Don’t get me wrong, ATSL has done a lot of good – but it can also be slow to adapt to new trends, and implement interventions in a dogmatic way instead of patient case based. Knowing ATLS is not an excuse to stop thinking.

This takes us to the case of the cervical collar. One main article to look at is a review article in the Scandinavian Journal of Trauma by Benger et al that points to four myths of the cervical collar:

1. Injured patients may have an unstable injury of the cervical spine.

2. Further movement of the cervical spine could cause additional damage to the spinal cord, over and above that already caused by the initial trauma itself.

3. The application of a semi-rigid cervical collar prevents potentially harmful movements of the cervical spine.

4. Immobilisation of the cervical spine is a relatively harmless measure, and can therefore be applied to a large number of patients with a relatively low risk of injury “as a precaution”.

Myth 1: Injured patients may have an unstable injury of the cervical spine.
Of course, trauma patients may have an unstable cervical spine injury. But the incidence seems to be low. In two studies on trauma patients who were considered at high risk of head and neck trauma, they found an incidence of 0,7% for significant cervical spine injury.

Myth 2: Further movement can cause additional injury.
Neurology can worsen after the initial accident, but is neck movement to blame? Edema and bleeding can surely attribute. But additional movement? In the awake patient, it seems like the patient will protect his own spine, just as he would protect a broken arm by automatically holding it still, and any tissue trauma and swelling will only help stabilise the area, even in the unconscious patient. And carefully moving the unconscious patient or carfully manipulating the neck is unlikely to cause more harm. In a patient cohort with confirmed cervical vertebral injury, 8% of them did not have their spine immobilised – but outcome did not differ.

In a study comparing the incidence of neck injuries in a first world country where cervical collars are applied, to a third world country without collars, there was no difference in the incidence of neurological injuries from the cervical spine. There are many flaws with such a study, but if we expected a big rise in secondary spine injuries, it should have been detected in these smallish samples.

Myth 3:  The cervical collar restricts neck movement.
This is one of the big ones. Everyone who’s worked with trauma patients know the collar doesn’t stabilise the neck much. Even a perfectly apllied collar allows for at least 30 degrees of flexion/extension/rotation.

OK, OK, so spinal injuries aren’t that common, cervical collars don’t really immobilise the neck, and collars have never been proven to affect clinical outcome – but how about we just put the collar on anyway, to be on the safe side… It sure couldn’t do any harm?

Myth 4: Just put on the collar anyway, to be on the safe side…
But is the cervical collar safe? Due to its pressure on the jugular veins and reduced venous return, it increases ICP. Not so good on an unconscious head trauma patient… But the patients with an actual cervical spine fracture – surely the collar will help these patients? In a cadaver study where they inflicted neck injury on the cadavers and then place a semi rigid collar around the neck, radiological studies showed that the collar increased the fracture crease by over 7 mm! The conclusion was that a cervical collar on these patients probably would have made the neck injury worse. Oops. Then there’s the airway management. The collar hampers airway management, and is also a big aspiration risk for a vomiting patient.

So, where’s the upside?
On the pro collar side, there doesn’t seem to be any evidence available to support the use of cervical collars, the upside seems to be largely theoretical. Benger thinks cervical collars should only be applied in an unconcious patient where you have a high degree of suspicion for cervical spine fracture. But one could easily make the argument that the cervical collar shouldn’t be used in these patients either – or in any other patients. Patients where you think cervical spine stabilisation is warranted, supporting the head with a few blankets or similar might be better – and will avoid the raised ICP and dislocated fracture.

The collar we put on millions of trauma patients every year has no proven benefit, no proven protection against secondary injuries. Still, we put it on with the A in ABC and take focus and time from more important interventions. For the patient with a high suspicion of spine injury, careful handling is needed – but not a cervical collar. For all other trauma patients, they will more than likely be better off without the collar. The HEMS service in Bergen has started delivering more and more trauma patients to the trauma bay without a cervical collar, the UK Faculty of Prehospital Care has recently released new guidelines for managing spinal injuries – and the counter-revolution has just started.

This post was put together from several articles and also much based on a Norwegian presentation and short article from Dr. Helge Asbjørnsen, consultant anaesthetist in the Norwegian HEMS. So a big thanks to my colleague Helge. His Norwegian article is linked in full text below.

Benger et al. Why do we put cervical collars on conscious trauma patients? SJTREM, 2009. Full text.

Prehospital use of cervical collars in trauma patients – a critical review, J Neurotrauma, 2013. Co-written by Helge Asbjørnsen.

Stone et al. The Effect of Rigid Cervical Collar on Internal Jugular Vein Dimensions, Acad Emerg Med, 2010.

Mobbs et al. Effect of cervical hard collar on intracranial pressure after head injury, ANZ J Surg, 2002.

Hauswald et al. Out-of-hospital Spinal Immobilization: Its Effect on Neurologic Injury, Acad Emerg Med, 1998. Full text.

Peleg et al. Extrication Collars Can Result in Abnormal Separation Between Vertebrae in the Presence of a Dissociative Injury, J Trauma, 2010.

Lador et al. Motion Within the Unstable Cervical Spine During Patient Maneuvering: The Neck Pivot-Shift Phenomen, J Trauma, 2011.

For Scandinavian readers, Dr. Helge Asbjørnsen’s article on cervical collars in Norwegian is a must:
“Har noen behov for – eller nytte av – nakkekrage?”

You might also want to read our post on PROGRESS IN PREHOSPITAL SPINAL INJURY MANAGEMENT

This entry was posted in Emergency Medicine, Prehospital Medicine, Trauma. Bookmark the permalink.

56 Responses to THE CURSE OF THE CERVICAL COLLAR

  1. LEMYZE says:

    Another common belief is that cervical collar is needed after attempted suicide by hanging. We recently described the case of a young guy who hanged himself in his garage. His neurological status dramatically deteriorated after admission to hospital until the cervical collar was removed (Lemyze M et al. Emerg Med J 2011; 28:532). Contrary to popular belief, hanging death is not commonly caused by asphyxia or spinal fracture but rather by blockage of the blood stream to the brain. Venous drainage from the head may be obstructed just by turning the head to one side or by applying a mass above 1.3 kg on the neck. Cervical collars that may exert an extrinsic compression on the vasculature of the neck should be absolutely avoided in this setting.
    Great blog! Thank you. Keep on this way!

    • Thomas D says:

      I read that case report – very interesting! And thanks for your comments!
      I suspect blood flow might even be compromised at lower pressures if the patient is bleeding and hypotensive.
      Your case report was one of the articles I read for writing this post, but in the end I had to shave off as much as possible to keep it short.
      The case report is well worth reading:
      “Unintentional strangulation by a cervical collar after attempted suicide by hanging”:
      http://emj.bmj.com/content/28/6/532.full.pdf

    • bogdan says:

      do a doppler in neck veins and arteries 24 hours later on that kind of patiet… i strongly support your oppinion

  2. Fantastic post thanks.
    You are preaching to the converted here. In my practice I am forever removing C-Spine collars from patients with nothing more than an isolated limb injury, bought in by Ambulance officers. When asked why the collar was applied the inevidable answer is either “it is part of our protocol” or “just to be sure” neither answer is adequite justification in my opinion.
    Bring on the revolution I say!

    • justin ambo says:

      that’s interesting ric. i recall fronting into your workplace many times and having my patient with no neck pain, tenderness or discomfort placed into a collar by the triage nurses. i have even had this happen when the patient has a fractured jaw…causing them great discomfort.
      it is also worth noting that Ambulance Paramedics operate under the medical directors licence … and so they are bound to stay within the wants and protocols/guidelines they require. most paramedics realise the use of collars is overkill, but are bound to the guidelines laid down by the director.

  3. Sue says:

    Thank you for this post – this is one of my big hobbyhorses! Cervical-spine immobilisation is an evidence-free zone. I have seen lots of unnecessary imaging in awake patients with minor injury mechanisms whose neck pain was mostly CAUSED by a collar. I’ve lost count of the number of times these people have sighed with relief when I’ve removed their collar on arrival.

    I’ve seen ambos immobilise a neck with NO pain, ”as a precaution”. You can only then conclude that, since their leg had no pain either, they should immobilise that as well. Why do people suspend intelligence in this setting? Is it something to do with the enormous PR machine that is ”the trauma system”? (Where no amount of imaging is too much?)

  4. Thomas D says:

    Thanks for your comments so far. I just want to chip in and say the cervical collar is worshipped by doctors as well as ambos. Many of our trauma team leaders will scorch ambo teams delivering trauma patients without a cervical collar, and stop all other interventions until the collar is in place. Not because it is needed, but because it’s in the hospital’s trauma protocol.

    So I know many ambos who put a collar on the patients where they don’t think it’s justified – just to satisfy the docs in the trauma bay. This is a discussion that needs to be had on all levels and all parts of the trauma chain.

    • Sean C says:

      Brilliant post and agree thoroughly I think a barrier to accepting that this is best practice for EMS systems out there is prob two main things one its always been the way we have done things and is seen to be best practice but is clearly now not and despite having a poor if any clear evidence base to be so it is still continued with, two the fear of litigation it is almost unthinkable that a cervical fracture may exist in a patient brought to hospital by Ambulance despite it being stable or otherwise so the answer to this is to just give every trauma patient one despite of the mechanism and of the clear emerging indications that it may cause more harm than good there needs to be across the board mindset change from all disciplines on this one so we can all use our intelligence on this and not just continue with the existing tick box way of managing these cases.

    • Sue says:

      Agree with your comments about some doctors being ”rigid” (pun intended) about collars too.

      We already have the data about safety of clinical clearance from NEXUS and the research behind the Canadian cx spine rules. What makes us ignore it?

  5. Preety George says:

    Very interesting article! I often apply C-spine collars on patients being transferred to the ‘better’ hospitals purely based on the fact that there will be a lot of flack to face if the pt dare arrive there without a collar on, even if the pt has been thoroughly assessed to have an isolated limb injury. (I unfortunately work in one of the poor hospitals that has no onsite CT imaging facilities). So I sympathise with the ambos who are forced to apply collars without any obvious indication! Wonder to what degree a broken leg would distract a patient from his broken neck…

  6. Chris W says:

    In the future please cite your sources in the text. Finding a style guide online is easy and will make for a much more professional and academic article. Also please proofread. Misspellings detract from any positive message you might have. If EMS is to gain credibility as a profession we need to go the extra mile and present appropriately.

    • Thomas D says:

      I appreciate your comments and candor, and your points are timely, but doesn’t necessarily fit with our view.

      Regarding references and form, part of the goal for this site is to avoid the formal, academic journal feel that makes you fall asleep. We lean more towards the blog format, it should be accessible and slightly opinionated – but still firmly based on evidence. And we feel we’ve found a good balance. For this post, however, with several referenced articles, I can see it might not be optimal – but this form has been our choice. The references are still there for anyone who wants to go in-depth on the subject, and a good start for further reading.

      Proofreading is done, but obviously not to perfection. Our main goal is to make sure the facts are correct and easily understandable, then comes spelling. Still, I strongly agree with the sentiment that misspellings are distracting. But what can I say – it’s a constant work in progress. We appreciate any corrections from our readers. But don’t throw the message out with the slightly misspelled bathwater.

    • Matt says:

      There’s spelling, grammar, etc errors in all medical resources, including those that pay companies to edit them. I find a response like yours insulting and more distracting than any misspelling or grammatical error. Comments like yours take away from the significance that this post makes.

      “Have no fear of perfection – you’ll never reach it.”
      ― Salvador Dalí

  7. Louise W says:

    Enjoyed the read. So your telling me that after having the misfortune of being rear-ended a while back, and being mandatorily provided with a semi-rigid collar which obliged me to lie directly on my injured lower back resulting in additonal back pain and new onset collar induced neck pain that there is insufficient evidence for that!

    Phooey to style guides, what needs some attention are current protocols and guidelines on how, when, and why we might use c-spine collars. And anyways, this blog is stylish enough.

    • Thomas D says:

      Well, the body of evidence in medicine is constantly changing, but yes, it’s quite clear that c-collars are at the very least heavily overused at this point in time. I’m sorry about your misadventures with the c-collar!

      I do like style guides, but we made our own. I’m glad you find it stylish.

  8. Josh says:

    Thanks for a terrific post.

    I work in a medical ICU, so I don’t have much experience with C-collars. However, we do admit a fair number of patients with syncope/seizure, and occasionally someone will think it’s a bright idea to slap on a C-collar. This invariably causes pain/agitation and (I think) increases sedative use thereby promoting delirium. It also makes management harder (i.e., femoral line placed due to inability to use IJ site), promotes unnecessary neck imaging, and distracts the medical team from what is actually going on.

    In sum, I feel that the C-collar *itself* is a distracting injury for medical patients.

    Best,
    Josh

    • Thomas D says:

      Thanks for your comment. I can relate to that description and totally agree with your assessment.

      And once the collar goes on, no-one has the guts to remove it without extensive testing and imaging.

  9. Pingback: Den där halskragen… | akut eftertanke

  10. Tom says:

    Hate to say it, but it’s a ‘cover our ass’ move. It’s seen as a standard procedure so, if a pt has neck injuries, we’ve just added our names to the list of defendants in a potential lawsuit if we don’t use it. Does it suck? Absolutely. But I’m not about to put myself on the stand and testify that I think hundreds of ER docs and medical directors are wrong, particularly as a meager EMT-B.

  11. Pingback: The LITFL Review 112 - Life in the Fast Lane medical education blog

  12. Rogue Medic says:

    Two papers showing some benefit from cervical collars –

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2691505/

    http://www.ncbi.nlm.nih.gov/pubmed/23079144

    The benefit was only a surrogate endpoint and only when backboards and KEDs were avoided.

    It appears that the benefit is due to prevention of the natural tendency to bend at the neck when exiting a vehicle. The collar does reduce that motion, but even with a collar, the backboard and KED produce more motion.

    .

    • Thomas D says:

      Well, it’s actually one study, and I don’t think it shows benefit for cervical collar in any meaningful way. I’ll explain:

      The two references are sort of the same study, the first one a pilot study, then the proper study. They use video motion capture to measure the degree of movement on extrication.

      My problem with the study is that they are using healthy volunteers. The point when moving/extricating CONSCIOUS patients WITH neck injury is that they will protect their own neck/spine. Just as a broken arm or leg, they will strongly resist moving their neck in any way that causes more pain/harm. That’s the theory, anyway 😉

      On unconscious patients, they won’t try to bend their head to facilitate extrication. Of course, the prehosp team need to make sure they stabilise the neck on extrication. A cervical collar still allows a good deal of movement, so you can’t trust it to keep the neck straight, you will have to have a person who watches and stabilises it.

  13. Ash says:

    If the clinical equipoise is there then the large rct should be done to decide this argument. In the mean time, we should stick to the current standard of care.

    • Thomas D says:

      I disagree. While we’re waiting for the big RCTs, I think the extra hassle and dangers of the cervical collar definitely means it should stay off – if we’re presuming clinical equipoise.

  14. Pingback: CME 12/12/13 Cervical Spine Immobilisation - Charlie's ED

  15. Pingback: BEST OF 2013 |

  16. He is right – if assume equipoise at best (rather than harm), then need to get rod of ’em now and then do trial…

    Of course, whether this will happen in medicolegal environment w/o leadership from ATLS or other bodies means hapless EMTs will have to continue putting ’em on…

  17. Rob Timmings says:

    Thomas,
    Sensational rogue you!!
    Brilliant blog style read, that gives real information that is worthy of reproduction, and I did on our page which generates some robust comentary. The wowsers who seem more intent on style guides and formatted referencing are missing the purpose of social media, which is to inform. It is first and formost a communication tool. And conversation is good communication (sorry I didn’t have references for that ).
    Well done on a great post. I’ll be looking for more from you. Rob.

  18. Pingback: CERVICAL COLLAR R.I.P. |

  19. Pingback: The Death of the Cervical Collar? | AmboFOAM

  20. Pingback: Collier cervical | thoracotomie

  21. Matt P says:

    Hi Thomas,

    Great blog mate, as a Paramedic I’ve been trying to wind back my use of unnecessary C- collars through use of NEXUS and/or Canadian C-Spine rules. Due to a number of times when I’ve been confronted by ED nurses and more often ED doctors who have then placed a collar and called into question my clinical abilities, I have lost some of the willingness to make these “rogue” decisions. Great to see more literature being promulgated to try and encourage dogmalysis.

    Matt

  22. Ron B. says:

    Finally streetwise medics are speaking up. It”s time to throw away the 1 Standard of Care fit all protocol. The over use of collars as well as backboards certainly has done more harm than good. The lead medic should have the right to decide if either or should be used. This CYA attitude and we’ll be liable in a lawsuit mentality needs to go.

  23. Les Murdie says:

    The really scary part is that quite often, patients only receive a cervical collar based upon the destination hospital….. i.e. “we’re going to Hospital ABC, we had better use a collar and headblocks” with NO consideration of what treatment is warranted.

    Scary, because we are allowing fear, not evidence to dictate patient management.

    And then there are the nurses and doctors who insist on forcing a patient to straighten their neck so a cervical collar can be applied, ignoring the pain and potential damage such manouevres can cause.
    If a patient has neck pain preventing them from moving their neck, then what is to be gained from forcing them to move their neck so a collar can be applied to (allegedly) immobilise the neck?

  24. Anthony says:

    “1. Injured patients may have an unstable injury of the cervical spine.”

    YOU CALL IT A MYTH, AND THEN YOU IMMEDIATELY STATE IN THE TEXT THAT IT’S TRUE. 0.7% That’s more than I (and I think many providers) thought. So, how is it a myth. You can’t start your article based on lies.

    Also you misspelled ATLS at one point. Proofread.

    • Thomas D says:

      Hi Anthony, thanks for your feedback. The myth in the first paragraph is that patients are believed to commonly have an unstable cervical fracture. That’s also what we also outline below. One thing is the 0.7%, another thing is the injury mechanism. Cervical collars and neck stabilisation has been implemented with the A in ABC almost regardless of injury mechanism and reported pain/injury by the patient. These are the things we address in no. 1.

      Regarding the 0.7%. Please read the full sentence: the 0.7% number was from two studies on trauma patients who were considered at HIGH RISK OF HEAD AND NECK TRAUMA. So even when we as health personnel consider the possibility of neck injury high, it works out to be around 0.7%. In the general trauma population, the number of unstable neck injuries is of course much lower.

      We do proofread. Mistakes slip through. So we appreciate readers pointing out errors – in spelling or otherwise.

  25. LW Osorio says:

    As a former military medic and current search and rescue operator I have seen along with others how many courses continue to make such a big deal about the c-collar and having people carry these in the wilderness. We are currently in the developing stages of a portable item that will enhance c-spine control while leaving the neck area open in order to continue the patient assessment. this is a great forum!

  26. Pingback: New kids on the block. 2015 APLS trauma updates. - St Emlyns

  27. MJ says:

    Hi Thomas

    Very valid point

    Been dealing with this issue for more than 20 years including war injuries.
    Understanding the natural protective mechanisms of the neck versus the pathological changes of injury is crucial.

    The mechanism of injury and any underlying pathology also to be part of the process.

    Possibly you can do updated search for the evidence and critically appraise relevant papers.

    LW Osorio would you pls explain about your new method of immobilisations?
    Thank you

    • LW Osorio says:

      The prototype is still in the developing stages for a mobile, lightweight product..not a c-collar that will enhance c-spine control in the wilderness or when help is delayed stage. The product will allow the front of the neck to continue to be exposed for further assessment and it would assist SAR teams or hike leaders to be able to carry it. More to follow since there are patents issues. ( Do not worry I did not take the idea from anyone, the developer is seating next to me)

  28. Pingback: Episode 26 – The Spinal Cord | FOAMcast

  29. Pingback: Prehospital Spinal Care | Clinical Focus

  30. Pingback: Backboard and Collar Nightmares : Emergency Medicine Cases

  31. Amelia says:

    Please forgive my ignorance on the topic, I am not a medical personnel. However, I am an individual recovering from a Type II Ondontoid Fracture who has been in an Aspen c collar for 8 weeks so far and still has 7 more weeks to go plus the possibility of a future fusion of my C1 and C2.
    So my question is in reference to myth 2. Are you telling me that there is no medical need for me to be in a collar and that I will naturally protect my neck from further damage? Because I have to be honest, being in a hard c collar 24/7 is an absolute nightmare. However, I will continue to follow my neurosurgeons advice because, well it is my neck and I’m lucky to still have use of all my limbs.
    Thank you for your blog but I need further evidenced based practices.

    • Thomas D says:

      There’s a huge difference between the acute management of trauma, and the healing process of your spine that you are going through. Basically, it needs to stay aligned just like a broken arm or leg would, to grow together in the best possible way. So in such a setting a collar would have a place. We can’t provide any medical advice here, but would strongly suggest you follow the recommendations of your neurosurgeon. Best of luck with your healing.

  32. Pingback: JC: Self extrication vs. assisted extrication. St.Emlyn's - St.Emlyn's

  33. Pingback: ATLS- och trauma-myter | Huddingeanestesin

  34. victoria says:

    Can anyone point me to the ‘evidence’ that led to us starting to use collars in the first place? Or has it always just been based on theory?
    Thanks

    • Rob Timmings says:

      The mandate of applying a collar in the “A” part of primary assessment was a corruption of the mantra ” Airway and simultaneous C-spine precautions”

      Those same c-spine precautions meant that in cases of trauma where index of suspicion for injury to the neck was present, clinicians should not use traditional airway gestures like head tilt. From there the leap was erroneously made to application of c-collars.
      It is , was and always be crazy to think that a clinician will dick around sizing me up for a collar when he/she hasn’t even checked if I’m breathing yet.

      For years I’ve taught precautions to mean, absence of unnecessary movement, not restrict movement.
      Rob.

  35. Neil Long says:

    Excuse my ignorance but is this only condemning the use of the hard collar or does this include applying a phili or aspen collar.
    Most of my patients get a phili while awaiting CT of their c-spine. If there is a fracture then all good. If they still have C-spine pain they go home in the phili and follow-up in the trauma clinic +/- MRI. Is this practice wrong according to this literature or are we only talking about the hard collars?

    • Thomas D says:

      Thanks for your comment. We’re reviewing the hard collar as used for the initial handling of trauma patients. Our concern has been that the collar takes focus away from the critical ABC interventions and might hinder some of them (this would include all sorts of collars), as well as introducing some side effects that may be bad for an ABC compromised patient. In an ABC stable patient at GCS 15, you can do pretty much whatever you want – also with regards to collars.

      One excemption would be Bechterew’s patients, where forcing their stiff neck into a “neutral” neck position in a collar could lead to harm.

      We’re doubtful about how much the phili/aspen collar will help the patient in your described setting, but there’s no real evidence that points to serious harm. A neurosurgeon would be better to answer to possible benefits of applying a collar in your example, outside the initial trauma setting.

  36. Angie says:

    Is your portable collar available yet? If so how do we get more information about it?

  37. Lee says:

    I am a nurse educator in an emergency department and I have read a lot of the evidence supporting the removal of hard collars from practice. Our department has removed the rigid collars only to be replaced with philli collars.
    I am struggling to find evidence supporting the use of philli collars and their clinical benefit. I receive the same patient complaints for the philli collar (discomfort, pain claustrophobia) as i did for the hard collar and their application seems to be a reason for clearing the spine not to be a department priority.
    Our department seems to be split 50/50 in the doctor management of suspected c-spine injury and the application of philli collars. However, I would say 90-100% of the nursing staff see them as causing harm and loath applying them. This puts most of us in a difficult position as it is completely doctor dependant can can lead to some heated discussions and we are overruled any way.

    • Rob Timmings says:

      Lee, I hear your lament, and as a Nurse Educator with over 18 years of experience in trauma education, I fully support your observation. This gap (time warp) transition from evidence to practice needs to be closed. Are we headed for anothe Oxygen and Chest pain debarcle with this Collar or no collar business? God I hope not.
      Dogmalysis needed.
      Rob (ECT4Health)

  38. Pingback: Spinal Motion Restriction – rathemblog

Leave a Reply

Your email address will not be published. Required fields are marked *